Smith ScholarWorks Smith ScholarWorks

Theses, Dissertations, and Projects

2014

Clinical perspectives on etiology, assessment, formulation and Clinical perspectives on etiology, assessment, formulation and

treatment of imaginary companions in adolescents with treatment of imaginary companions in adolescents with

attachment trauma attachment trauma

Kathryn M. Collins Smith College

Follow this and additional works at: https://scholarworks.smith.edu/theses

Part of the Social and Behavioral Sciences Commons

Recommended Citation Recommended Citation Collins, Kathryn M., "Clinical perspectives on etiology, assessment, formulation and treatment of imaginary companions in adolescents with attachment trauma" (2014). Masters Thesis, Smith College, Northampton, MA. https://scholarworks.smith.edu/theses/761

This Masters Thesis has been accepted for inclusion in Theses, Dissertations, and Projects by an authorized administrator of Smith ScholarWorks. For more information, please contact scholarworks@smith.edu.

Kathryn Collins Clinical Perspectives on Etiology, Assessment, Formulation and Treatment of Imaginary Companions in Adolescents with Histories of Attachment Trauma

ABSTRACT

This theoretical study explored the phenomenon of imaginary companions as they present

within the lives of adolescents with histories of attachment trauma. The phenomenon and is

origins were explored through a review of developmental, psychoanalytic and trauma research.

Theoretical perspectives of narrative therapy and Winnicottian object relations were then

introduced as lenses through which to conceptualize assessment and formulation of the

phenomenon, with careful consideration paid to the social context within which the phenomenon

emerges. These theoretical perspectives were then applied to a discussion of assessment,

formulation and treatment within a specific case example, written by Proskauer, Barsh and

Johnson (1980), about an Navajo adolescent male who presented for treatment with imaginary

companions and a complex history of trauma.

The study findings indicated that imaginary companions can be viewed as a resilient

adaptive response to attachment trauma and that therapy can assist adolescents in finding a place

of holding through which to address unmet developmental needs as well as to create multiple

storied identities. Both theoretical approaches focused treatment on the development of the

adolescent’s whole and true self. Areas of further research and the relevance of this study to the

field of social work were both explored throughout the body of the study.

Keywords: imaginary companion, attachment trauma, narrative therapy, Winnicott

CLINICAL PERSPECTIVES ON ETIOLOGY, ASSESSMENT, FORMULATION AND

TREATMENT OF IMAGINARY COMPANIONS IN ADOLESCENTS WITH

HISTORIES OF ATTACHMENT TRAUMA

A project based upon an independent investigation, submitted in partial fulfillment of the requirements for the degree of Master of Social Work.

Kathryn Collins

Smith College School for Social Work

Northampton, Massachusetts 01063

2014

ii

ACKNOWLEDGEMENTS

This thesis is dedicated to a memorable young person who shared her history and imaginary companion with me, inspiring me to learn more about the power of resiliency in the face of life’s most challenging and horrendous obstacles.

I am grateful for my mentors throughout this process including my supportive and loving

parents, my thesis advisor Kelly Mandarino, my second year internship supervisors Christine Dresp and Beth Muccini and the many wise and wonderful Smith School of Social Work Students who have gone before me, leaving helpful tips and inspiring legacies in their wake. The classrooms at Smith have been rich places of learning and growth and I thank the many professors who have awakened my mind and spirit, opening it up to new ideas and critical lenses through which to see myself, those around me and the world.

The long days of this thesis process were made lighter by loving family, friends, and my

roommates, Meghan and Lucia. They were also made cozier by small cafes, comfortable writing nooks and the Honey Pot, a place to call home. I am grateful for the running and biking trails of Boston that provided a beautiful landscape for blowing off steam. It is thanks to all of these spaces that the writing process could evolve and take root in me. Finally I am grateful for my partner, Julien, who helped me carry my load many times when I was feeling weighted down. For the laughter, adventure and dreaming that came along with that support, I am forever grateful.

iii

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...................................................................................................... ii TABLE OF CONTENTS .......................................................................................................... iii CHAPTER I INTRODUCTION ............................................................................................................ 1 II METHODOLOGY ........................................................................................................... 6 III PHENOMENON .............................................................................................................. 13 IV THEORY CHAPTER I: NARRATIVE THEORY .......................................................... 39 V THEORY CHAPTER II: WINNICOTT’S OBJECT RELATIONS ................................ 55 VI DISCUSSION ................................................................................................................... 68 REFERENCES .........................................................................................................................103

1

CHAPTER I

INTRODUCTION

It is not uncommon for children of preschool age to have imaginary friends, imaginary

companions and invisible friends. Multiple studies have shown that this type of fantasy

production is relatively common in early childhood and preschool aged children (Gleason, 2004;

Singer & Singer, 1990; Taylor, Shawber & Mannering, 1999; Taylor & Carlson, 1997; Trionfi &

Reese, 2009). Research by Dierker, Davis and Sanders (1995) suggests that approximately one

third of children between the ages of 2 and 10 develop such companions. Furthermore, according

to Taylor, Hulette, and Dishion (2010), such fantasy production has been associated with positive

characteristics such as, “advanced theory of mind, narrative skills, and the ability to get along

well with others” (p. 1632). According to Davies (2011), “the appeal of such play is that it

supports the school-age child’s wish to be competent and powerful, even in the face of extreme

odds” (p. 350). Early psychoanalysts surmised that thirteen to twenty percent of children go

through a phase in which they play with an imaginary companion and most children who

experience the accompaniment of an imaginary friend are between the ages of three and six years

old (Sperling, 1954; Svedson, 1934; Hurlock & Bernstein, 1932). More recent studies have

focused on empirical data and have found that, by the age of 7 years old, about 37% of children

have had an invisible friend (Taylor, Carlson, Maring, Gerow & Carley, 2004). However, the

imaginary companion phenomenon is not restricted to this age group. Harriman (1937) found

that even some college students had maintained relationships with imaginary friends from

2

childhood. The present study is interested in adolescents who experience imaginary friends long

after the typical developmental period. More specifically, this study is interested in how

imaginary companions take shape in the lives of adolescents for whom experiences of trauma in

early caregiving relationships have interrupted normal development.

For the purposes of this study, the terms imaginary companions, invisible friends and

imaginary friends will be used interchangeably as they have often been used as such in research.

The literature on imaginary companions that has been published can largely be classified into

two basic areas: normal child developmental literature (Taylor, 1999) and psychopathology

literature (Putnam, 1997; Sawa, Oae, Abiru, Ogawa, & Takahashi, 2004). Researchers have

sought to understand the phenomenon through conducting empirical research, both qualitative

(Seiffge-Krenke, 1997) and quantitative (Taylor, Hueltte & Dishion, 2010), as well as through

psychoanalytic studies (Bach, 1971; Benson & Pryor, 1973; Sperling, 1954), focusing on the

presentation of imaginary companions in typically developing children. Furthermore, much of

the research has attempted to provide a framework for understanding imaginary companions,

seeking to label them as positive or negative, adaptive or maladaptive, and pathological or

healthy (Brooks & Knowles, 1982; Friedberg, 1995; Harter & Chao, 1992; Meyer & Tuber,

1989; Sanders, 1992; Seiffre-Krenke, 1997; Somers, 1984; Trujillo, Lewis, Yeager, & Gidlow,

1996) Elizabeth Kandall (1997) captured trends in literature when she wrote,

A child’s relationship with an imaginary companion has been understood as a positive sign of intelligence, creativity, social ability, maturity, and also as a negative sign of social/academic incompetence, propensity to live in fantasy, isolation, loneliness and a prodromal sign of multiple personality (p.1)

A small body of research has sought to explore the phenomenon of imaginary companions within

adolescents and adults (Allen, 2004, Harriman, 1937) and adolescents who have experienced

trauma (Gupta & Desai, 2006; Adamo, 2004). Much of the psychoanalytic research regarding

3

imaginary companions is older (Bach, 1971; Benson & Pryor, 1973; Sperling, 1954) and relies

heavily on case studies of patients with significant psychopathology. There remains a large

research gap at the intersection of imaginary companions, attachment trauma and treatment as

few researches have approached the phenomenon through a broad lens of formulation and

treatment. Furthermore, discussions of culture, class, ability, sexuality, race, ethnicity,

spirituality and gender as they pertain to the phenomenon have largely been left out of the

research all together maintaining a focus on intrapsychic phenomenon without proper

consideration for the historical context, present reality and dynamics within the environment of

the person presenting with the imaginary companions.

This study will conduct a theoretical inquiry into this phenomenon through the lenses of

Winnicottian object relations and Narrative Therapy. It will engage in inquiry around etiology,

resilience, ecology, pathology, as well as consider the implications for therapeutic treatment. I

will then review a case study of an adolescent with a history of attachment trauma and imaginary

companionship. Through this practice, I will flush out the theoretical concepts and ground them

in a way that will fill gaps within the existing literature and contribute to the wealth of

knowledge within the field of clinical social work.

Study Questions

Most of the scientific research dealing with ICs has been concerned with preschool aged

children who develop imaginary companions as playmates in response to “situations of special

stress and traumatic character” (Nagera, 1969, p. 192). The normative developmental literature

has regarded imaginary companions as “fairly common” for typically developing children,

sharing that imaginary companion’s are useful in “aiding the child in the developmental process

in a transitory, compensatory manner” (McLewin & Muller, 2006, p. 532). The present study

4

will try to answer the following questions: How do imaginary companions play a role in the

development of adolescents with histories of attachment trauma? Are imaginary companions in

adolescence an indicator of pathology or resilience? What can we learn from science, research

and theory with regards to how to treat adolescents who present with imaginary companions and

a history of attachment trauma? How can Winnicott’s object relations theories and Narrative

Therapy theories be used to conceptualize treatment of adolescents with histories of attachment

trauma and imaginary companions? What do clinicians need to keep in mind while working with

adolescents with histories of attachment trauma and imaginary companions? This study hopes to

expand upon the current research and engage a conversation about treatment modalities that

could be helpful for this population. Secondly, this study hopes to draw conclusions that can

serve to inform practice interventions within the field of clinical social work. Ultimately, this

study holds hope that its findings could improve treatment for these clients and support a shift

away from a pathology-focused understanding of the phenomenon, enhancing the body of

literature available with a well researched and practice based approach to treatment. Ultimately,

an inquiry such as this is necessary for the field of social work which endeavors to consider

clinical treatment of person in environment, expanding the knowledge base for assessment and

treatment to support the work that social workers do in addressing individuals as well as the

systems and environments they exist within.

This study will involve an extensive literature review, considering both empirical and

psychoanalytic research that has been written about imaginary friends. The research will cover

developmental research with broad strokes and focus more specifically on research that is

particularly relevant to the phenomenon.

5

The following chapter on methodology will introduce key terms and concepts to ground

the forthcoming study. The chapter will provide an overview of the structure of the study

including a summary of the foundational perspectives that will be offered within the phenomena

chapter as well theories that will be discussed in the two theory chapters. In addition, the chapter

will introduce the reader to the study’s teaching goals, the methodological biases of the

researcher, and the strengths and limitations of the present study.

6

CHAPTER II

METHODOLOGY

“And then I would hold up my hand and move every finger and each joint of my arm; and see

the shadow answering, answering, answering. And I should nod to it and bow to it and curtsy.

And I would dance to my shadow alone. And all this I would do again and again all day….”

–Lord Dunsany, Charwoman’s shadow, 1926

This methodology chapter will introduce the reader to key terms and concepts that

ground the present study’s inquiry into the phenomenon of imaginary friends in adolescents with

a history of attachment trauma as well as to the structure and format of the study. The reader will

also be introduced to the study’s teaching goals, the methodological biases of the researcher, and

the strengths and limitations of the present study.

Key Terms and Concepts

The first widely read articles on imaginary companions was wrote by Svedson (1934)

who defined an imaginary companion as:

An invisible character, named and referred to in conversation with other persons or played with directly for a period of time, at least several months, having an air of reality for the child but no apparent objective basis. This excludes that type of imaginative play in which an object is personified, or in which the child himself assumes the role of some person in his environment.

(p. 988)

Svedson’s (1934) description has endured and been used as the primary description for

imaginary companions in much of present day research. The present study is particularly

concerned with the experience of such companions in adolescents with histories of attachment

7

trauma. The adolescent period of development, ages from 12 to 18 years old, is late for what is

considered the “normal” time frame for a person to make use of an imaginary companion

(Somers & Yawkey, 1984). Additionally, this study is concerned with a particular area of

traumatic experience termed, attachment trauma, which shall be understood as traumatic

experiences such as physical abuse, sexual abuse, verbal abuse and physical and emotional

neglect that occur within the context of child-caregiver relationships (Allen, 2011). These acts of

commission or omission are traumatic in that they result in harm, potential harm, or threat of

harm to the child (Center for Disease Control, 2012). For the purposes of this study, attachment

trauma should be understood as occurring in relationships between children and their parents or

primary adult caretaking figures in the child’s life. Allen (2011) writes that, “attachment trauma

is especially detrimental because it undermines the primary function of attachment, which is to

provide protection” (p. 20). Furthermore, Allen (2011) writes, “attachment trauma is the most

extreme example of the failure of good enough caregiving, because it not only constitutes a

failure of protection but also places the child in danger” (p. 21). It is this researchers belief that

imaginary companions are often closely linked to this particular kind of trauma and that

imaginary companions are the result of resilient adaptations within a child’s internal working

model for attachment. Early on in life, attachment figures are often a parent or parent substitute;

however, attachment relationships are understood to play out with other attachment figures

throughout the life cycle “(e.g. beyond a mother figure to a father, alternative caregivers,

siblings, extended family, friends, partners, spouses, and even groups)” (Allen, 2011, p. 21).

These important attachment relationships will also be considered, as they are relevant to the time

period of adolescence explored through the present study. The field of trauma research is broad

and there has been much written about attachment trauma specifically. The present study will be

8

limited to understanding attachment trauma as it relates to the neurobiological, psychological and

social underpinnings of the imaginary companion phenomenon.

For the purposes of this study, the terms imaginary companion, fantasy friend and

imaginary friend will be used interchangeably and researched as equal concepts. As stated

previously, imaginary companions are to be understood as invisible name bearing persons who

present in the psychic world of the constructor and communicate directly with the constructor in

a mutual way over a period of time (Svedson, 1934). They exist in the realm of imagination,

however are different from a separate identity or self. Thus, the phenomenon of a child who

takes on an imaginary character or calls him/herself by another name during play would not meet

criteria. The companion must be invisible to anyone expect the child who is creating it. Hence,

physical objects such as stuffed animals or toys imbued with imaginary characteristics and

persona would not fit the criteria for this study.

Flow and Structure

The following study will explore the phenomenon of imaginary companions through

multiple lenses; conceptual, theoretical and experiential. Within the phenomenon chapter of this

study, the reader will be introduced to literature and research from the fields of developmental

psychology, psychoanalysis and trauma as they relate to the phenomenon of imaginary

companions within the lives of children. The chapter will focus its efforts on understanding how

imaginary companions manifest in the lives of adolescents, considering the impact of trauma on

the developing brain. Finally, the phenomenon chapter will introduce the clinical case of Ben,

written by Stephen Proskauer, Elizabeth T. Barsh and Lusita B. Johnson in 1980. Ben, a Navajo

adolescent boy of 17, presented to treatment with an imaginary companion and a significant

9

history of attachment trauma. The case explores both his history and treatment experiences

through modality of hypnotherapy (Proskauer, Barsh & Johnson, 1980).

Later in the two theoretical chapters, the reader will be introduced to two theories

relevant to the study of imaginary companions in the lives of adolescents who have experienced

attachment trauma: narrative therapy theory and Winnicott’s object relations theories. The

narrative therapy theory chapter will provide a history and context for narrative therapy and

thereafter explore two narrative therapy practices that are particularly relevant for addressing the

phenomenon in treatment: externalizing conversations and re-authoring conversations. The

chapter on Winnicott’s object relations theories will review a history of Donald W. Winnicott

and his theories, focusing specifically on three concepts offered by Winnicott: the concept of the

good-enough mother, the concept of transitional objects and the concept of the capacity to be

alone. Thereafter, the reader will be primed to consider how therapists from these two schools of

theory and practice might think about and approach the phenomenon of imaginary friends in

adolescents who have experienced attachment trauma.

Finally, the case of Ben will be explored in-depth within the discussion chapter,

grounding the concepts and theories presented within this study, and engaging an integrative

dialog around the ways assessment and theory translate in to treatment. The theories and

concepts presented in the early chapters of this study will be discussed and integrated through

critical analysis of the case and phenomenon therein; in this way, the author and the reader will

be equipped to consider the phenomenon in all of its complexity. The discussion chapter will

also offer ideas and practical tools to consider for treatment that apply to the case of Ben as well

as globally to the phenomenon as it presents in adolescence. Ultimately, through reviewing the

10

present study, the reader will be equipped for assessing and treating adolescents who present to

therapy with imaginary companions and a history of attachment trauma.

As a researcher, I am biased in choosing these two theories. I admire the way that they

capture resilience. I have chosen theories that conceptualize the source of intrapsychic problems

broadly, encompassing relational and environmental influences as well as components of

individual psychology. Within my own clinical work, I have encountered two adolescents with

histories of attachment trauma and the companionship of imaginary friends. My short work with

these young people was challenging, confusing and thought provoking. Through the work I

realized that I could not be of much help if I did not have a clearer understanding of the

phenomenon. I questioned both resilience and pathology when I met with these young people

and wondered what I could offer as a therapist to shore up their strengths, provide a positive

attachment experience, and help them to navigate the confusing aftermath of childhood

experiences that were riddled with abuse and neglect at the hands of caregivers. I also wondered

about the intersection of this phenomenon and culture and wanted to know more about how

imaginary companions could be interpreted through cultural lenses other than those of western

medicine and science. I wanted to take a step back form the conversations about pathology that

took place at the agency where I was a social work intern and, in exchange, I wanted to delve

into a thesis that could shift the perspective on this phenomenon to one of resilience.

While the present study will offer a focused look at theory, treatment and practice in an

area that has been relatively unexplored, the study is not without its limitations. The case

material was not written with this study in mind and hence, valuable insights about Ben’s history

and presentation will be missing in the case summary. Furthermore, Ben had undergone

treatment by the time the study was written, thus the treatment and the study itself were colored

11

by the biases of the researchers who wrote the study as well as by the theoretical orientation and

treatment approach they used therein. Finally, the field would benefit from more empirical

research, both qualitative and quantitative, regarding the experiences of children whose

experiences are in line with this phenomenon. As a field, we still have much to learn about the

impact of attachment trauma on children’s fantasy lives including children’s utility for and

attachment to their imaginary companions in the aftermath of trauma. The present study will

represent a move in a curious direction but will make only a dent in the knowledge tank; we still

have much to learn from children and the imaginary friends they create.

Much of the psychoanalytic research that has discussed the phenomenon through the use

of clinical case study in children older (A. Freud, 1936; Sperling, 1954; Bach, 1971), similarly so

with adolescents (Benson, 1980; Klein, 1985). Furthermore, the psychoanalytic tradition’s

exhaustive use of clinical case study to research the phenomenon of imaginary friends may have

contributed to their frequent associations with psychopathology (Manosevitz, Prentice, &

Wilson, 1973). Until a study by Sawa, et. al. in 2004, there had been no research with regards to

the function of imaginary companions in the actual psychotherapeutic process. One strength of

the present study is that it will consider the phenomenon through two modalities: clinical case

study and discussion of theory and treatment. Furthermore, the study endeavors to apply a lens of

resilience, which could balance out the research associating the phenomenon with

psychopathology. Finally, the study’s theoretical orientation creates space to discuss and analyze

different ideas, a practice that inherently sheds light on theoretical and treatment strengths and

biases and brings the discussion of this phenomenon to a level of complexity that is unparalleled

within the current research.

12

Through this study, I expect to find new ways of interpreting and understanding the

phenomenon. I also expect to identify and provide recommendations to therapist who encounters

this phenomenon in clinical contexts. Finally, I expect to complicate the pathological lens

through which the phenomenon has commonly been understood and formulate a

multidisciplinary understanding that captures the resilient nature of the phenomenon. The

following chapter begins this journey by introducing psychoanalytic and psychological research

that will frame the discussion of imaginary companions going forward.

13

CHAPTER II

PHENOMENON

“Never trust a friend who deserts you in a pinch.”

–Aesop: The Two Fellows and a Bear

The following chapter will explore psychoanalytic writing and psychological research in

order to construct a scaffolding framework for understanding the phenomenon of imaginary

companions in the lives of older children with histories of attachment trauma. Through

examination and synthesis of research from the areas of developmental psychology, psychiatry

and psychoanalytics, this chapter will explore different lenses through which theorists and

researchers have understood the role of imaginary companions within the lives of children.

Furthermore, the chapter will present and summarize some of the research that has been done at

the intersection of imaginary companions, older children and trauma. There will be a section on

how imaginary companions take shape in the lives of children as well as a section delving into

the impact of trauma on the developing brain and developing self. The trauma section will

introduce readers to the concept of traumatic dissociation and consider the resilience of the

imagination as a tool for coping with distress. Research gaps will also be discussed. Finally, this

chapter will introduce and summarize a case written by Stephen Proskauer, Elizabeth T. Barsh

and Lucita B. Johnson about a Navajo adolescent male named Ben, who presented to treatment

with an imaginary companion and a history of complex trauma.

14

Developmental Lens

Developmental research suggests that imaginary companions are normally temporary,

transient phenomena, appearing first in early childhood (Inuzuka, Satoh & Wada, 1991; Pearson,

Rouse, Doswell, Ainsworth, Dawson, Simms, Edwards & Falconbridge, 2001; Somers &

Yawkey, 1984; Taylor, 1999), and disappearing by late childhood for most children (Inuzuka et

al., 1991; Taylor, 1999). The prevalence of imaginary companions drops considerably between

the ages of eight and twelve years (Pearson et al., 2001).

In the middle to late 1900s, theorists began to discuss imaginary companions as a normal

part of cognitive development, and even as a sign of superior intelligence (Fein, 1975; Nagera,

1969). They surmised that the child who develops an imaginary companion is also showing the

capacity for advanced cognitive structures such as the capacity for symbolic thought, abstract

thinking, and elaboration (Fein, 1975). Piaget (1962) believed that imaginative play provided the

child with an opportunity to symbolically play out experiences and assimilate reality (i.e.,

parental moral authority) and did not believe that this type of play was a defensive process of

denying and avoiding reality. Instead, he believed that imaginary companions could contribute to

a child's creative growth by encouraging him/her to develop cognitive elaboration and

originality. Piaget also viewed imaginary companions as helpful to children as they learned to

explore the environment, develop new skills, and cope with emotions.

Jersild, Jersild and Markey (1933) studied children ages 5-12 with imaginary companions

and found that a greater number of girls were able to give definite descriptions of imagined

playmates and that a higher IQ was correlated with this ability to make definite descriptions.

Furthermore, Svendsen (1934), in a study of 40 cases of children with imaginary friends, found

that they were encountered three times as often among girls as boys (Bender & Vogel, 1941).

15

Both Sperling (1954) and Nagera (1969) contend that children use companions to aid through a

specific developmental stage, “the imaginary companion frequently plays a specific role in the

development of the child and once that role is fulfilled, it tends to disappear and is finally

covered by the usual infantile amnesia” (Nagera, 166).

While the developmental research serves to broaden the framework for understanding

imaginary companions, it does so through samples of normally developing children and does not

speak to the experiences of children who have experienced trauma or who present with

imaginary companions later in development.

Psychoanalytic literature has taken a somewhat broader focus, although much of this

literature was written over 40 years ago. For instance, Nagera (1969) wrote about the similarities

and differences between the imaginary companions of younger (2 – 8 years old) and older

(latency aged) children. He believed that for children 2 to 8 years old, imaginary companions

served a particular function in terms of supporting development; this function was different than

the role played by imaginary companions of older children. He saw the imaginary companions of

young children as intermediate steps between controlling forces from the outside (i.e. their

parents) and their own sense of internal control (i.e. “superego structure”). In contrast, he

observed that older children use companions as a way of coping with impulses that seem to

escape their control due to frustration, stress or conflicts. For both older and younger children,

these imaginary companions serve the function of controlling the child’s behavior, yet in

different ways given the child’s developmental capacity for self-control (superego function)

(Nagera, 1969).

Some psychoanalytic research, such as Philip Harriman’s 1937 study, has engaged in

inquiry about the presence of imaginary companions later than the typical developmental period.

16

Harriman studied college students who presented with imaginary friends, he found evidence for

the assertion that more “older persons” experience this type of phantasy production than prior

research has indicated. Harriman (1937) says, “chiefly impressive is the apparent fact that an

imaginary friend is an illustration of wishful thinking compensatory for a real or fancied

deprivation of completely satisfying flesh-and-blood companions” (p. 370). Additionally, he

found that there was no apparent harm to the individual as a result of such fantasy production

and that it was different that the vengeful and erotic phantasies that might be found in an

individual with autistic thinking and a low capacity for imaginative play (Harriman, 1937).

Furthermore, Bender and Vogel (1941) wrote up their psychoanalytic observations of 14

children they saw within the Children’s Ward of Bellevue Psychiatric Hospital. In their study,

they observed children with a range of behavioral and emotional problems, yet with a common

history of attachment issues and trauma. They found that some children create imaginary

companions in a constructive way, “making up for deficient reality” (Bender & Vogel, 1941, p.

59). They found that others experienced imaginary companions in their phantasy life as a way to

internally mend brokenness within their family system. Furthermore, children created imaginary

friends that were scapegoats for the child’s defects, personifications of the hated parts of their

caregivers and vehicles for overcoming parental rejection. All in all, the researchers concluded

that these imaginary companions were all constructive for the child and valuable in the

development of their personality. For most of these children, the imaginary companions made up

for deficits within the child’s world. Bender and Vogel (1941) found that two particular factors

contributed most to these phantasies: “1) an unsatisfactory parent child relationship” and “2)

unsatisfactory experiences from the world of reality due to unfavorable social or economic

situations” (p. 64). They regarded the phenomenon as not only normal but resilient, a way for the

17

children to create a more integrated personality to deal with the conflicts in their lives. Adamo

(2004) echoed some of these findings when he wrote, “socially withdrawn adolescents and

adolescents with a higher level of distress are at high risk of creating fantasy friends” (p. 150).

There is strong feeling in both of these cases that imaginary friends in adolescence are an internal

response to an environmental deficiency. When the adolescent’s world is lacking in sufficient

attachments, the imaginary friend can present as an auxiliary support to bolster feelings of social

isolation and disconnection.

Bender and Vogel (1941) reference the 1929 work of Kirkpatrick who asserted that

nearly all children have imaginary companions at some point in their lives and that sometimes,

“the imaginary companion is an ideal self, sometimes a scapegrace, and at other times it is not

the self at all but a distinct personality (p. 57). This raises the question: what form do these

imaginary companions take?

Qualities of Imaginary Friends

For as long as researchers have been writing about imaginary companions, they have

explored the shape taken by these invisible companions in the lives of children. What do these

friends look like? How do they function? Does the child believe them to be real or are they aware

that the imaginary friend is a figment of their imagination? This section will review common

descriptions of imaginary companions, as well as discuss what research says about the roles they

play in the lives of children and adolescents.

Perhaps the most often cited definition of imaginary companions was offered by

Svendsen (1934) who described an imaginary companion as, “an invisible, name bearing person

who presents for the constructor a psychic reality over an extended period of time and whom he

can refer to in his every-day life communication” (p. 985). Hurlock and Burnstein (1932)

18

believed that, “these playmates are not merely vivid ideas, or imaginings, but actual visual and

auditory projections. They can be seen and heard as vividly as if they were living children” (p.

381). Piaget (1951), a well regarded developmental psychologist and philosopher observed his

own three year old daughter playing with her imaginary friend and wrote, “this strange creature,

which engaged her attention for about two months was of help in all that she learned or devised,

gave her moral encouragement in obeying orders and consoled her when she was unhappy, Then

it disappeared” (p. 130).

In a 1933 study, Jersild, Jersild and Markey examined how imaginary companions took

shape within the imaginations of children ages 5 to 12 years old. Of the 143 children that

reported imaginary companions, 79 percent of the imaginary companions took the form of

human beings, story characters or elves and fairies. The 21 percent remaining were represented

by animals, dolls and special objects (Jersild, Jersild & Markey, 1933). Svedsen (1934) believed

that imaginary companions were accompanied by strong visual imagery. Harvey (1918) believed

that children always knew that their imaginary companion was a figment of their imagination,

understanding imaginary companions were visual or auditory ideas that become as vivid and real

as a visual or auditory percept (Nagera, 1969). Nagera (1969) observed that imaginary

companions were often the same age as the child, or slightly younger, but that they were never

an adult figure. Although they never took the shape of adults, Nagera believed that imaginary

companions could possess adult characteristics such as, “strength, power, knowledge, authority,

etc.” (Nagera, 1969, p. 171). Bender and Vogel (1941) observed that, “the child who has created

an imaginary companion enters into active physical play with its creation. He is not content to

play in phantasy, but carries out the same activities and plays the same games as with a real

playmate (p. 64).

19

Sawa, Oae, Abiru, Ogawa and Takahashi (2004) differentiated between the way that

children experience imaginary companions and the way that adolescents experience them saying,

“adolescents experience them with a sense of reality and think of them as other persons, while

also recognizing that they are not real” (p. 145). Sawa et. al. argue that, because adolescents can

acknowledge that their imaginary companions are not real, this phenomenon should not be

classified as pathological when presenting in adolescence. They found that in adolescents,

imaginary companions provided advice, acceptance, encouragement, and an outlet for repressed

aggression. In each of the three cases they explored, the imaginary companion functioned to

fulfill an ability that each patient desired but could not act on alone. Adamo (2004) observed that

the imaginary companion of his adolescent client, Salvo, served as a form of relationship

fulfillment that his client could not find out in the world. Salvo was observed to be able to

visualize a vivid image of his imaginary friends and use willpower to call to them when they

were needed (Adamo, 2004). The research is limited and there remains a great deal to learn about

the shape that imaginary friends take in the lives of adolescents.

It became clear through this literature review that the research on imaginary companions

has been minimal and that there is still a great deal that we do not know about children and their

imaginary companions. The research is especially minimal with regards to imaginary

companions in the lives of adolescents with histories of trauma. McLewin and Muller (2006)

suggest that future research should include case studies stating, “case studies are able to

accumulate in-depth information regarding their clients that is generally not gathered in

empirical studies” (p. 542). I experienced great difficulty finding case studies to work with for

this study and when I did begin to find them, I discovered that much of the writing was decades

old and written by authors who were primarily interested in the psychopathology of the patients

20

they wrote about. It is difficult to know exactly why the research has been so minimal; however,

I believe this could be due, in part, to the reality that children and adolescents often conceal the

reality of their imaginary companions and do not discuss them with their caregivers, family or

friends. Jersild (1968) found that almost all children invent imaginary companions, yet they go

mostly undetected by their parents. Psychoanalytic studies have found that imaginary

companions can served a compensatory function when children are faced with a loss or a deficit

in close relationships with caregivers; however, in no case was this the reason for referral

(McLewin & Muller, 2006). In order for an imaginary companion to enter the conversation

between the constructor and the adults in their lives, there must be trust, openness, and

communication. For many children with histories of trauma, the capacity for trust has been

damaged and their imaginary companions may result from the lack of close relationships with

caregivers. Hence, researched may have moved away from this phenomenon because, due to its

very private nature, it is difficult to study and may soon disappear with the onset of safe and

trusting relationships.

Pathological Lens

Early theorists and clinicians who wrote about imaginary companions were divided as to

whether they were developmentally normal or a sign of pathology. Psychoanalytic writers of the

mid-1990s primarily wrote about imaginary companions through case studies, focusing on the

psychopathology of disturbed children. More contemporary research has discussed imaginary

companions as a form of dissociation for children with the most extreme form of dissociation

resulting in a dissociative identity disorder (DID) (McLewin & Muller, 2006). DID has been

described as the most extreme response to psychological trauma (Putnam, 1997) and manifests as

two or more distinct personality or identity states, called alters, which recurrently take control

21

over a person (Pica, 1999). A large body of research links DID to a history of child abuse

(Putnam, Guroff, Silberman, Barban, & Post, 1986; Schultz, Braun, & Kluft, 1989), The same

research indicates that most individuals are not diagnosed until they reach their 20s. Pica (1999)

asserts that for individuals with DID, the dissociative defense serves an adaptive function,

protecting the child from overwhelming psychological trauma; however, reliance on this defense

creates problems within social, occupational and familial relationships.

Some psychoanalytic research regarding DID indicates that there is a “window of

vulnerability”, in development between the ages of 3 and 8 years old (Putnam, 1995). Marmer

(1991) attributed this vulnerability to the child’s tendency to use the defense of splitting at this

time in addition to their underdeveloped understanding of the difference between self and object.

Furthermore, Allison and Schwartz (1980) found that the age during which the child was

traumatized is significant, finding that those who experienced trauma before the age of 6 showed

greater disorganization and a greater number of alter personalities, while children who were 8 or

older exhibited fewer personalities and greater ego strength. It is during this time that,

developmentally, the child uses preoperational and concrete operational cognitive reasoning,

meaning that they see the world in rigid, black and white terms. It is also during this time in

development that children often begin to develop imaginary companions and are generally more

hypnotizable than during other periods of development. Pica (1999) drew the conclusion that

children use this natural “hypnotic capacity” to dissociate in response to psychosocial stressors.

Furthermore, he believed that a child’s experience in infancy, before the “window of

vulnerability,” was significant. He asserted that, experiences early on in development could

predispose a child to increased vulnerability during the aforementioned window of time (Pica,

1999).

22

In concert with Pica’s ideas, Fink (1988) suggested that DID represents a disorganization

of the child’s core self; the foundations of the self that a child creates during the first 6 months of

infancy. In describing children with core self disorganization, Fink viewed that they “lack

control over their bodies, experience internal fragmentation, have trouble integrating emotional

experiences, and have a tendency to confuse the past with present” (Pica, 1999). Furthermore,

with this theory in mind, a child will be more vulnerable to developing DID when they are both

predisposed by experience of vulnerable attachments in infancy as well as victims of traumatic

incidents within the vulnerable window of development; a time which would allow alter

personalities to develop.

In the absence of a soothing caregiver, Pica (1999) believed that these children were

much more likely to rely on dissociative defenses and find it difficult, if not impossible to tell the

difference between phantasy and reality. Pica (1999) says, “without knowing the conditions

under which these children live, teachers, babysitters, and neighbors are likely to see these kids

as moody, hyperactive, shy and withdrawn when really they may be experiencing continuous

dissociation, drifting in ad out of fantasy depending on the stress they experience in the

environment” (p. 407).

Wickes (1966) was interested in the difference between normal and pathological use of

imaginary companions and suggested that, in “normal cases” imaginary companions serve a

transient function and are disposed of when the child no longer needs them. In this way, they

serve a function in supporting the child to navigate something new, challenging, or lonely. In

contrast, Wickes found that children who make pathological use of an imaginary companion use

them as a way to escape reality. In this way, the pathological use of imaginary companions often

produces a split between the child’s thinking self and the self of feeling and relationships. The

23

imaginary friend is a personification of the things that the child cannot own, accept or tolerate as

coming from or happening to the self (Wickes, 1966).

When imaginary friends are encountered in the play of older school aged children and

adolescents, psychoanalytic tradition has often interpreted these companions as a concerning

indicator of psychopathology and an indicator of a child’s underdeveloped capacity to blend

fragmented parts of themselves (Gupta & Desai, 2006; Friedberg, 1995) or a way to defensively

split off a memory, feeling or part of the self (Taylor, 1999). In the early 1900’s, psychoanalytic

writers primarily used case studies of disturbed children to draw conclusions about imaginary

companions. Early writing conceptualized imaginary companions as defensive in nature,

allowing a child to cope indirectly with something unpleasant or overwhelming (Freud, 1968;

Nagera, 1969; Sperling, 1954). While it was acceptable for young children to have imaginary

companions, it was considered indicative of a mental disorder as development progressed

(Svendson, 1934).

As the aforementioned research suggests, imaginary companions could be interpreted as a

form of cognitive coping, which is beneficial for older children in stressful or traumatic contexts

(Compas, Hinden, & Gerhardt, 1995). It is also possible to imagine that these imaginary friends

could be a form of entertainment that is not associated with dissociation, resilience or pathology.

They could be fulfilling developmental changes in “reality-testing, fantasies, creativity and

changes in socio-emotional development and in friendship conceptions” (Gupta & Desai, 2006).

Overall, the presence of these fantasy friends in the lives of older children remain poorly

understood (Seiffge-Krenke, 1997) and as a result, can represent an important but unexplored

area for clinical social workers and other mental health service providers to consider.

24

Trauma and Dissociation

A final layer to add to this literature review is one of exploring the phenomenon of

imaginary friends in older children through the lens of trauma. Trauma refers to the, “enduring

and adverse impact of extremely stressful events” (Allen, 2011, p. 4). Allen (2011) writes, “the

essence of trauma is feeling terrified and alone” (p. 4) and it is this experience of fear, terror and

loneliness that can leave a damaging and lasting psychic effect. Along the spectrum of traumatic

experiences, impersonal trauma is at one end and attachment trauma is at the other. Relatively

impersonal trauma is exemplified by natural disasters such as earthquakes, floods and fires.

These events pose a physical threat and danger to the self and others and can pose a significant

risk to psychopathology. However, at the other end of the spectrum is interpersonal trauma,

defined as, “deliberate threat or injury in the context of interpersonal interaction” (Allen, 2011, p.

13). This is the most problematic kind of trauma that an individual can experience. Allen (2011)

writes, “interpersonal trauma is a major contributor to severity of trauma, and attachment trauma

is the worst” (p. 5). Attachment trauma occurs within the contexts of child-caregiver

relationships and represents trauma not only to the relationship between child and caregiver but

also to the attachment system itself. Allen (2011) writes, “attachment trauma damages the safety-

regulating system and undermines the traumatized person’s capacity to use relationships to

establish a feeling of security” (p. 22). The present study is interested in attachment trauma and

its biological, psychological and social impact on the developing child.

Individuals with attachment trauma are unable to access memories of a time before their

trauma. These individuals are challenged with the prospect of revisiting stabile cognitive

structures given their inadequate exposure to the modeling necessary to develop these cognitive

structures. Allen (2001) states that, “openness to intense emotions and a capacity to wrestle

25

actively and deliberately with complex meaning – reflective function – is required for growth”

(p. 344). In expounding upon the impact of attachment trauma, Allen (2001) quotes LaMothe

(1999) writing, “there is a horrifying void of an obliged, trustworthy, and faithful other and in

this void there is consequently the absence of psychological organization and the very structures

of subjectivity […] these experiences cannot be organized because the very symbols which

human beings depend upon for psychological organization cannot represent the very absence of

obligation, trust, and fidelity” (p. 344). Given the early onset of attachment trauma and, by its

very nature, the absence of a safe attachment relationship, persons who have experienced

attachment trauma are often rendered defenseless of cognitive structures and the capacity for

emotional regulation that would be present in the minds and bodies of non-traumatized

individuals (Allen, 2011).

Applying a biopsychosocial lens to assessment and treatment of trauma is crucial due to

the marked neurobiological and evolutionary impacts of trauma. Allen (2011) suggests that we

can hardly begin to treat trauma without understanding the impacts of both biology and

evolution. This section will elaborate on the impact of trauma on the brain and will also speak to

the impact of trauma on a person’s self-concept and relationship to the world and others. In

speaking of the biological perspective Allen (2011) says, “the lens of evolution enables us to

grasp, most fully that, no matter how pathological they may seem, our clients’ trauma symptoms

often can be best understood as adaptive efforts” (p. 9). Furthermore, from a developmental

psychopathology lens Allen (2011) explains, “attachment trauma not only generates extreme

distress but also, more importantly, undermines the development of mental and interpersonal

capacities needed to regulate that distress” (p.10). McLewin and Muller (2006) indicate that

child maltreatment represents a significant risk factor for maladaptive development and function

26

writing, “samples of abused children exhibit increased difficulties in negotiating state-salient

developmental issues, including the development of secure attachment relationship, the

development of autonomy and a sense of self, and the development of adaptive strategies for

regulating behavior and affect (p. 540).

Thus, therapeutic work intended to address attachment trauma is complex and requires

time, creative approaches, and attention to non-verbal treatment such as art, music and

movement therapies, and attention to the manifestation of the trauma response within the body.

The key to helpful therapy is that it must occur within a trusting and safe relationship between

the therapist and the client. Such a relationship will enable the mind to think and feel about the

trauma in different ways and gain access to emotional refueling vis-à-vis the trusting

relationship. In this way, clients are able to transition from painful reliving to a process of

remembering; a stance that involves a greater sense of personal agency and distance than was felt

during the trauma itself. It is also crucial that the therapist understand the unique way that a

client’s history of trauma has influenced his/her framework for relationships as well as the

defensive coping mechanisms used to protect the self. For an individual that presents with an

attachment trauma history and the presence of imaginary companions, trauma research suggests

that it is important to consider this type of fantasy production in terms of the defensive coping

mechanism of dissociation (Allen, 2001).

The literature on imaginary companions indicates that there is some form of dissociation

involved in the creation of imaginary companions (McLewin & Muller, 2006). Dissociation

refers to a variety of behaviors associated with lapses in psychobiological and cognitive

processing (Ogawa, Sroufe, Weinfield, Carlson, & Egeland, 1997). The Diagnostic and

Statistical Manual (DSM- IV-TR) defines dissociation as “a disruption in the usually integrated

27

functions of consciousness, memory, identity, or perception” (American Psychiatric Association,

2000 p. 519). Dissociation is considered to be a developmentally normal response to

psychological stress, and is often helpful for children; however, it may represent pathology if it

persists as a primary coping mechanism as development progresses (Ogawa, Sroufe, Weinfield,

Carlson & Egeland, 1997; Shirar, 1996).

Dissociation often persists as a primary coping mechanism for individuals who have

experienced trauma and can undermine functioning in several ways such as impacting memory,

compromising cognitive functioning, blocking coping mechanisms and in the most extreme form

resulting in “alters” or dissociative identities and the DSM-IV diagnosis of Dissociative Identity

Disorder (Allen, 2001). Studies have consistently documented the association between

dissociation and a history of severe sexual abuse (Kisiel & Lyons, 2001; Giolas & Sanders,

1991), and noted trauma histories in almost all individuals with dissociative disorders (Briere &

Runtz, 1988; Hornstein & Putnam, 1996). As dissociative disorders have been linked to the

presence of imaginary companions, this research is particularly applicable to the present study

and is one trauma lens through which the phenomena can be explored.

Research has conceptualized children's imaginary play, including the creation of

imaginary companions, as a mild form of dissociation (McElroy, 1992; Shirar, 1996).

Dissociative characteristics are more evident in children than they are in adults given children’s

use of play and fantasy activities (McLewin & Muller, 2006). In creating an imaginary

companion, a child may defensively split off a memory, feeling, or aspect of self (Taylor, 1999).

McElroy (1992) suggests that if a child chronically relies on imaginary companions in this way,

they may eventually take on full-fledged personalities; meeting criteria for a diagnosis of

Dissociative Identity Disorder.

28

There is a wealth of research in the field that considers the impact of trauma on the child

brain. Schore (2001) found that early trauma alters the development of the infant’s right brain,

the hemisphere that is responsible for processing social and emotional information, bodily states,

and attachment. Bashman (2008) suggests that, “as children protect themselves from the ravages

of cumulative relational trauma, they then retreat both psychologically (through dissociation) and

neurobiologically (by overly restricting their yearnings for attachment)” (p. 432). Schore (2001),

a prolific infant researcher, believes that caregiver-induced trauma contributes to more frequent

and more intense psychopathology in children. Early childhood trauma lays down a way of

relating where the infant withdraws into their internal world and dissociates (Bashman, 2008).

On a neurological level, stress triggers the production of adrenaline and cortisol, two

chemicals that the body naturally produces in order to cope with stress. The release of adrenaline

is essential to survival in that it mobilizes energy during a stressful situation and alters blood

flow to allow the body to effectively deal with stress. Cortisol also mobilizes energy in addition

to suppressing immune response, helping the body to cope effectively with high levels of stress.

When these hormonal systems are activated frequently, as is often the case for children who are

exposed to circumstances of chronic abuse and neglect, serious developmental consequences are

possible. Long term elevated cortisol levels have been linked to actual architectural changes

within the hippocampus, the part of the brain that is essential for learning, memory, and stress

regulation. Additionally, research has shown that excessive and chronic stress can alter gene

expression, meaning that a chronically stressed mother will be more likely to give birth to

children who are predisposed to fearfulness and reactivity when faced with stress (National

Scientific Counsel on the Developing Child, 2009). Teicher (2005) writes that, “such abuse, it

seems, induces a cascade of molecular and neurobiological effects that irreversibly alter neural

29

development” (p. 1). Teicher (2005) suggests that child abuse leaves lasting scars that are quite

difficult to treat later in life due to the physical alteration that take place within the brain .

Teicher’s (2005) concluding comments are powerful:

Society reaps what it sows in the way it nurtures its children. Stress sculpts the brain to exhibit various antisocial, though adaptive, behaviors. Whether it comes in the form of physical, emotional or sexual trauma or through exposure to warfare, famine or pestilence, stress can set off a ripple of hormonal changes that permanently wire a child’s brain to cope with a malevolent world. Through this chain of events, violence and abuse pass from generation to generation as well as from one society to the next. Our stark conclusion is that we see the need to do much more to ensure that child abuse does not happen in the first place, because once these key brain alterations occur, there may be no going back

(p. 8)

In conjunction with what research shows about the adverse effects of a stressful environment,

research has also found that a supportive and caring environment in early infancy can alter the

architecture of the brain in a positive way, changing some of the regulatory components of the

stress system. Hence, the relationship that children have with caregivers early in their lives is

critical in the development of the child’s ability to manage stress on a neurobiological level

(National Scientific Counsel on the Developing Child, 2009).

There has been some research to suggest that individuals who have experienced trauma

are more likely to retreat into fantasy in childhood to defensively cope with trauma, neglect and

isolation (Allen, 2001). Allen (2011) calls these individuals, “fantasy-prone” and writes that

they, “populate their worlds with imaginary companions, and they endow their dolls and stuffed

animals with feelings and personalities” (p. 129). Furthermore, Allen indicates that fantasy

production can be a sign that an individual has experienced trauma and noticing this behavior in

an individual can be important assessment information for therapists (Allen, 2011). Allen frames

this fantasy production as a form of dissociation, which the person detaches from that which was

traumatizing and turns inward towards fantasy production in a way that is pleasurable rather than

30

distressing. Allen (2011) writes that, “absorption in the inner world preclude engagement and

coping with the outer world” adding that, “for clients with dissociate disturbance and PTSD,

absorption in the inner world can be a slippery slope” (p. 178). This kind of internal immersion

renders the person vulnerable to post –traumatic flashbacks and more severe dissociative states

(Allen, 2011)

Researchers have made the connection between trauma and dissociation as well as

dissociation and the presence of imaginary companions, however, the presence of imaginary

friends in the lives of older children remains poorly researched (Taylor, Hulette & Dishion,

2010; Seiffge-Krenke, 1997) and even less research is available with regards to older children

who have experienced trauma (Gupta & Desai, 2006). As a result, this is an important, yet

unexplored area for clinical social workers and other mental health service providers to consider.

More research needs to be done in this area in order to best serve and bolster the resiliency of

older children that present for therapy with imaginary companions.

Research Gaps

Early psychoanalytic thinkers such a Hurlock and Burstein (1932), Sperling (1954),

Vostrovsky (1895), Bender and Vogel (1941), and Harriman (1937) were interested in learning

about when imaginary companions typically emerged for children, finding through their research

that children of a variety of ages could experience imaginary companions. Somers and Yawkey

(1984) found that the developmental course of this type of pretend play is not well understood,

but concluded that imaginary companions are not solely a phenomenon of early childhood, as

has often been supposed. However, other researchers disagree. For instance, Gupta and Desai

(2006) state that the phenomenon of an imaginary companions is “usually regarded as normal in

31

children, but if encountered in adolescence, it indicates some underlying psychopathology”

(p.149).

Some researchers comment on the strengths of a young person who experiences an

imaginary friend and others comment on the pathological structures of the phenomenon. In this

way, there are gaps in the research, disagreements among researchers, theorists and scientists

who have studied this phenomenon over the past 100 years. Furthermore, the psychoanalytic

tradition’s exhaustive use of clinical case study to research the phenomenon of imaginary friends

may have contributed to their frequent associations with psychopathology (Manosevitz, Prentice,

& Wilson, 1973). Until the study by Sawa, et. al. in 2004, there had been no research with

regards to the function of imaginary companions in the actual psychotherapeutic process.

While research into the connection between imaginary companions and dissociation is

illustrative of the dynamics of this phenomenon, there remain major gaps in our understanding.

While research has linked trauma and dissociation as well as dissociation and imaginary

companions, it had done little to link the three phenomenon together.

Discussing attached-related trauma, Bowlby (1980) described “segregated systems” of

defenses in which children separate painful emotion from consciousness in order to cope. Such

thinking is consistent with the understanding of dissociation as a survival mechanism in the face

of overwhelming trauma. By way of this understanding, the self is developed by making

meaning from experiences, and integrating these meanings and experiences over time. If the self

cannot gain meaning and make sense of experience, integration cannot occur (Ogawa et al.,

1997). Thus, if experiences, or aspects of experience, are dissociated often enough,

fragmentation of the self may occur. Over time, these fragments can become more distinct and

unavailable to consciousness (Peterson, 1991). Such theories of early development and trauma

32

are suggestive of the real threat that caregiver abuse and neglect can pose for the developing

child (McLewin & Muller, 2006).

Blizard (1997, 2003) discusses the impossible dilemma a child experiences when faced

with frightening or abusive behavior from a caregiver. Children must protect themselves from

abusive caregivers while still maintaining a relationship with the caregiver whom they need for

survival. The child may thus develop separate “good” and “bad” representations, or working

models, of the self, and “nurturing” and “abusive” models of the caregiver. These multiple,

incompatible models may become dissociated, and interfere with integrative functions of

memory, consciousness, and identity (Liotti, 1999). Over time, these representations may

become elaborated into multiple self-states, or alter personalities (Blizard, 2003; Liotti, 2004).

There is currently little crossover between research investigating childhood imaginary

companions as part of normal development, and childhood imaginary companions in children

with various forms of psychopathology (McLewin & Muller, 2006). However, Blizzard’s

conceptualization of “good” and “bad” selves as well as “nurturing” and “abusive” models of the

caregiver is an appropriate segue to discussing how theory may be useful in developing a greater

understanding of this phenomena.

Amidst all of the literature in the field related to conceptualizing and understanding

imaginary companions, I was not able to find any literature around how to treat children with

imaginary companions outside of in-depth case studies (eg. Adamo, 2004; Gupta & Desai, 2006;

Benson, 1980; Prosakauer, Barsh & Johnson, 1980). From what is apparent, this is a huge

research gap in the field. Although I continued to expand my search criteria, I found few studies

that provided an in-depth application of theory to the phenomenon of imaginary friends.

Furthermore, few articles have considered the treatment implications when a clinician utilizes

33

assessment and diagnosis to treat clients with imaginary companions through one modality or

another. The present study will focus in this way, diversifying the information available on the

phenomenon, creating new questions to consider, and broadening the discussion of the

phenomenon.

The Case:

In 1980, Prosakauer, Barsh and Johnson wrote a research paper entitled, “Imaginary

Companions and Spiritual Allies of Three Navajo Adolescents” regarding their experiences in

treatment with three Navajo adolescents at the Tuba City Hospital Mental Health Unit over a two

year period form 1971-1973. The authors were interested in the unique connection between

Navajo culture and the imaginary companion presentation, writing, “Navajo culture has woven

into its very fabric a tendency to dissociative phenomenon, exemplified by the nature and

influence of witchcraft beliefs as well as the prevalence of hysterical seizures” (p. 153). When

speaking of the jarring adjustments forced on Navajo children at this time, Prosakauer, Barsh and

Johnson (1980) write:

The child of a traditional Navajo family lives in an isolated encampment, often with maternal aunts, uncles, cousins and grandparents as well as his own immediate family. Extended family members play a correspondingly large part of his life, particularly his maternal uncles and grandparents. Group cooperation is essential to survival in a marginal sheep-herding economy. Though economically poor for the most part, Navajo families are culturally rich by virtue of their subtle language and complex religious healing ceremonies which require total community involvement and combine the functions of sacrament, medical treatment, and social gathering. At the age of five to seven, many Navajo children are abruptly removed from their families to attend a government boarding school, and environment that could hardly be more different from the family life they have known previously (p. 154).

Navajo adolescents that were sent to white-man boarding school went through a number of

challenging transitions in order to adjust to the school setting and then back again to their

communities of origin. These transitions represented a complex web of psychosocial conflict.

34

The authors say that, “the resulting conflict of cultural influences added to their difficulties not

only in coming to terms with personal spiritual power but also in resolving past emotional

traumata, accepting sexual maturation, and forming a unified identity” (p. 154). Ben was a

Navajo adolescent in the midst of transitioning back to his home and family for the summer after

a year in boarding school when his parents brought him to the Tuba City Mental Health Unit. His

case will be described and summarized here and used again in the discussion portion of this

study to ground the theories and research discussed herein.

Ben was the oldest of three children born to his mother and father. Ben’s mother had

been pregnant four other times, yet all the others had died in infancy. One of his infant siblings

suffered from severe malnutrition, a contributing factor to the cause of death. The authors

described Ben’s mother as a “domineering hypomanic” and his father as “schizophrenic”

(Prosakauer, Barsh & Johnson, 1980, p. 162). Ben’s father served during World War II and

suffered a serious breakdown. After her returned from war, he used peyote and alcohol

excessively to manage psychological distress and, on several occasions, became psychotic,

allegedly committing many assaults within the community including murder and rape. He made

a took up silver working and Ben worked with his father when he came home for the summers.

Ben’s father was feared within the community and shunned by angry relatives. Ben once

witnessed an attempt by his father’s relatives to push Ben’s father out of a moving truck; Ben

was two or three years old at the time. Ben’s father underwent multiple psychiatric

hospitalizations, each occasion receiving the diagnosis of schizophrenia (Prosakauer, Barsh &

Johnson, 1980).

Ben left for a white man’s boarding school at the age of seven, returning to his family

every summer. By age 17, Ben was performing as a top student, was an active member of the

35

debate team, and had plan to attend college and move onto a career in computer programming.

Prosakauer, Barsh and Johnson (1980) write that, “Ben gave the impression of being a

handsome, articulate, highly intelligent, scholastically successful and psychologically well-

organized adolescent” (p. 162). As an adolescent, Ben was aware of his family’s shameful

reputation within the community and felt a strong sense of obligation to them. He was awarded

with a college-credit tour of Europe, yet instead of going on the summer trip, he returned home

to spend the summer with his family. During that time, Ben’s father reportedly became acutely

psychotic, assaulting Bens’ mother and trying to strangle Ben at a Christian revival meeting.

Ben’s father was hospitalized after this assault and at the same time, Ben left home for a

orthopedic surgery to fix an injury in his shoulder. Ben and his father returned home from their

respective time away at the same time and resumed their silver work. Two days after their

return, Ben awoke in the night, screaming from a nightmare. The morning after the nightmare,

Ben was awoke speechless. His parents took him to the Tuba City Mental Health Clinic where

Ben met with a clinician and reportedly, “made a drawing of the black female figure which had

threatened to kill him in his dream” (p. 163). The clinician hypnotized Ben and under hypnosis,

Ben reportedly, “spoke in the whispered voice of a spirit companion dwelling inside of him” (p.

162). Prosakauer, Barsh and Johnson (1980) write:

Ben’s companion identified himself as the spirit of an Indian youth who had lived 300 years before. He had dwelt in Ben’s body since Ben had been a small child and it was he who had now stopped Ben from speaking. The spirit companion explained that the black figure in the dream was a second spirit, an enemy Spaniard girl, trying to kill Ben in order to prevent Ben’s companion spirit from reaching the end of his centuries-long quest. The companion spirit was friendly to Ben but rendered him mute because the evil spirit could strike him dead at once if Ben should speak (p. 163).

Ben’s companion gave an account of his history sharing that he had lived long ago and, during

that time, had conflict with the evil Spaniard girl. Per the companion’s story, the Spaniard girl

36

had tried to kill the companion at one point. Ben’s companion spirit escaped the attempt at

murder, however, his close friend died instead. When Ben’s companion spirit later died at the

hands of whites and Spaniards who had chased the Indians, he vowed to inhabit the bodies of

living Indians in an effort to find his old friend again (Prosakauer, Barsh & Johnson, 1980).

Prosakauer, Barsh and Johnson (1980) write, “the companion spirit described his

relationship with Ben as teacher, guide, and close friend ever since he first came when Ben was

two years old” (p. 164). Furthermore, “the evil spirit was with him, too. Ben tried to get rid of

the evil one by going to church and helping people, but the evil one would not let him behave as

well as he would have liked” (p. 164). The companion spirit described Ben as shy, quiet, truthful

and honest. The companion also spoke to Ben’s fear of closeness when he indicated a belief

through Ben that when people get too close, they die. The companion spirit blamed this dynamic

on the evil spirit inhabiting Ben, saying that if she were gone, Ben would not fear getting close to

people and he would be able to make friends. The companion declared that Ben, his family and

his girlfriend must go to a certain spot in order to be free of both the good male companion spirit

and the evil female spirit; saying that until then, Ben would remain unable to speak (Prosakauer,

Barsh & Johnson, 1980).

Ben’s companion spirit spoke through him in the third person, praising Ben’s goodness

and strength and blaming his faults on the evil spirit. Ben engaged in daily intensive

hypnotherapy. During one session, Ben said:

I had a friend who taught me things, who I had fun with. He was always with me wherever I went. We shared our feelings and fears and everything together. When I felt low I used to go out by myself, used to talk to myself and it seemed like it would talk back to me. I don’t feel that way anymore. It feels like I’ve lost somebody, part of me or something like that – like a real close friend, someone you’ve had since you were a baby, like a brother (Prosakauer, Barsh & Johnson, 1980, p. 165).

37

In describing his feelings after the companion spirit left him, Ben said, “in bed, I’d talk to myself

and from somewhere deep inside it seemed he was answering all the questions I’d ask. Now, it

feels like when you take the engine out of a car – can’t even remember a time when my friend

wasn’t there” (p. 165). While the companion was described as within him, the female companion

was described as acting on him, trying to rid him of his companion friend (Prosakauer, Barsh &

Johnson, 1980).

Growing up, Ben felt close to his uncle and his uncle was the only person Ben told about

his experiences of his imaginary companion before his companion was made known in the

mental health facility. Ben’s uncle taught him about spirits and was a mentor for him with

regards to other areas of life. Prosakauer, Barsh and Johnson (1980) quote Ben as saying, “I

always turned to him. I loved him and I was never scared of him.” It was the authors’ belief that

Ben’s companion, in ways, represented his uncle and that the loss of his imaginary companion in

therapy brought up unresolved grief about the loss of his uncle who died two years prior to Ben’s

treatment in the clinic.

During his treatment at the clinic, there were times when Ben described a conflict within

him. For instance, while describing the behavior of lying to his girlfriend he said, “I don’t mean

to do it. It seems like something was fighting inside me, one telling me not do it, the other saying

to go ahead. I had very little control over the fight” (p. 165-166). He also said that it “seems like

a battle going on inside me, conscience on one side and on the other I’m all mixed up. It seems

like its all part of myself” (p.166). He described the support given to him by his companions

saying, “when I was ten, my father was in the hospital in California, I didn’t know where to go. I

wanted to drink but my friend told me all the reasons not to, so I’ve never tasted liquor, never

smoked, because my friend stopped me” (p. 166). He described the lessons he learned from his

38

friend saying, “being free is being able to love and to give everything you have to another

person, instead of trying to satisfy yourself. To suffer for others is the best freedom there is” (p.

166).

The authors indicate that, at first, Ben believed his friend to be real, not imaginary.

Furthermore, when he was rid of the companion, he actively grieved the loss. As treatment

progressed, Ben was able to put words to fears of connecting with others and was increasingly

able to engage in real and connected relationships. Finally, the authors indicate the Ben

reconnected with his Navajo roots and decided to go to college to study to become a silversmith.

He reported that he still felt the presence of his friend at times, often while doing a Navajo craft

or doing an Indian dance. During these moments, he reportedly tried to recall how his companion

would have instructed him in the past. Furthermore, he reflected on the utility of his imaginary

companion during his time at the white man’s school describing his companion as a teacher in

instructing Ben how to hold onto his pain and create something with it. Ben committed himself

to creating and crafting long into his adult life (Prosakauer, Barsh & Johnson, 1980).

With the case of Ben in mind, this study will now transition to an introduction of theory.

Aspects of Narrative Theory and Winnicott’s theory of object relations will be used to review the

case in this study’s discussion. A history of narrative theory will be presented first, then the

following chapter will elaborate on the two narrative concepts of externalizing conversations and

re-authoring conversations.

39

CHAPTER IV

NARRATIVE THEORY

“To make our own meanings out of our myriad stories is to achieve balance – at once a way to be part of and apart from our

families, a way of holding and letting go”

– Elizabeth Stone (1988, p. 244)

This chapter will focus on the theoretical underpinnings of narrative therapy, providing a

framework for how narrative therapy might be used to work with an adolescent presenting with a

history of attachment trauma and an imaginary companion. The chapter will start by describing a

brief history of narrative therapy, introducing the postmodern perspective that grounds narrative

theory and practice. Next, the chapter will introduce two foundational practices used by narrative

therapists: externalizing conversations and re-authoring conversations. By way of this

introduction, the reader will be properly oriented to the narrative therapy lens, which will be used

to discuss how a narrative therapist might approach and explore treatment within the case of Ben.

Narrative therapy is founded on the belief that our experiences of life and self are

profoundly influenced by the stories we tell and the stories told about us (Madsen, 2007).

Narrative therapy was developed in the early 1980’s by two clinicians, Michael White and David

Epston, who were trained as social workers and family therapists and practiced at the Dulwich

Center in Adelaide, Australia (Goldenberg & Goldenberg, 2013). White and Epston integrated

ideas from philosophy and anthropology while developing their ideas, eventually weaving in

concepts from feminist thought as well (Monk, 1997). Their vision was for a therapy that is

collaborative and non-pathologizing; an approach to counseling and community work that is a

40

co-creative process between the client and the therapist; a therapy that believes people’s behavior

cannot be understood apart from their unique, systemic, sociocultural contexts (Smith, 1997, p.

31).

Gregory Bateson, an anthropologist and psychologist, was an early inspiration to White

and Epston. Bateson developed concepts related to the subjective nature of reality and the nature

of learning (Monk, 1997). White and Epston were drawn to Bateson’s concept, “news of

difference,” which suggested that, “in order to be able to detect and acquire new information,

human beings must engage in a process of comparison, in which they distinguish between one

set of events in time and another” (Monk, 1997, p. 7). Michael White built on this idea,

discovering that by drawing clients’ attention to subtle changes related to the growth and

reduction of their problems, he could support them to develop new insights into their abilities

and, in turn, help them to develop clearer perspective with regard to addressing their concerns.

Further fine-tuning occurred out of dialog between Michael White, David Epston and Cheryl

White during which they discussed their understanding of the “story metaphor”; the way people

use stories to make sense of their experiences. In relation to the story metaphor, White has

suggested that the narratives we construct about our lives do not capture the full richness of our

lived experience, however they do have an impact in shaping our lives. Thus, White

conceptualized a narrative therapy through which the counselor is encouraged to listen carefully

for events and occurrences outside of the dominant problem narrative, building these small

asides into new story lines that can take root in the person’s life outside of the dominant problem

narrative (Monk, 1997).

Narrative Therapy was also strongly influenced by the ideas of French historian and

philosopher Michael Foucault who wrote about the subjugation process that occurs within

41

professional practice. In Foucault’s writing he, “discusses how society constructs ‘true’ standards

of behavior, with which individuals feel obliged to comply” (Monk, 1997, p.8). Narrative

therapists avoid generalizations about normal and abnormal standards of behavior due to the

influence of Foucault’s observations about the abusive power of dominant discourses. Nicholas

and Schwartz (2008) quote Foucault as saying:

Too often in human history, judgments made by people in power have been imposed on those who have no voice. Families were judged to be healthy or unhealthy depending on their fit with the ideal normative standards. With their bias hidden behind a cloak of science and religion, these conceptions became refined and internalized. One-size-fits-all standards have pathologized difference due to gender, cultural and ethnic background, sexual orientation and socioeconomic status

(p. 294).

White developed these ideas and applied them to therapeutic practices believing that the

establishment of standards within therapeutic professions positions the therapist in the role of,

“classifying, judging and determining what is a desirable, appropriate or acceptable way of life”

(Monk, 1997, p. 8). In contrast, narrative therapists assume that knowledge is socially

constructed and that there are many diverse ways of understanding others and ourselves.

Following postmodern thought, these ideas are culturally informed and situated in ever-changing

local environments and relationships (Smith, 1997).

Narrative therapists believe that the process of creating new stories, outside of stories

about problems that people often bring to therapy, can help clients to actively change and

reshape their lives (Goldenberg & Goldenberg, p. 395). White and Epston define narrative

therapy as a process by which therapists “work collaboratively with people in identifying those

ways of speaking about their lives that contribute to a sense of personal agency, and that

contribute to the experience of being an authority on one’s life” (White, 1995, p. 121). When

people have lost power, as is the case when abuse has taken place, they have often done so by

42

way of learning to understand themselves from the perspective of others or from the perspectives

of the problems in their lives. Roberts (1994) suggests that, “as people tap into the power of

stories and restore to flow of meaning between past, present and future, they begin to access

memory and imagination; they find their voice, break silenced stories, elaborate minimal stories,

open rigid tales, and change vantage points from which to tell and understand stories” (p. xiii).

This process can restore the sense of “personal agency” that White (1995) says can become lost

beneath the veil of social construction as well as traditional individual and systemically oriented

language for understanding people and problems. The process of restoring personal agency is

central to the aim of narrative therapy.

Broader Context:

A cornerstone of narrative thinking and narrative therapy is the foundation it holds in

“postmodern” thinking; meaning the contention that knowledge is socially or consensually

constructed (Smith, 1997, p. 5). This postmodern thinking is grounded in the observation that our

social world is like a fabric, tightly woven by threads of belief about what is acceptable and

unacceptable, as well as with ideas about which ideas and people are relevant and worth

following. Postmodern thinking observes that this fabric is shaped by those with power and is

difficult to unravel, yet important to deconstruct through thoughtful inquiry and examination of

social norms. Social construction is particularly relevant to the field of therapy in that traditional

therapies have reinforced social construction, allowing therapist to use a cloak of science while

considering diagnosis and theory, approaching their clients through the lens of pathology (Smith,

1997, p. 5-6). Paul Dell (1982) suggests that this practice reflects bias and differing values:

Pathology exists in the domain of human intentions and values – not in the domain of science or in the domain of absolute reality. When we imply that a person is objectively and absolutely sick, we are confusing these two domains and losing track of the value of judgments we are making. Similarly, and more

43

importantly, when we try to “fix” the person who (we imply) is objectively and absolutely malfunctioning, we are again confusing our values with absolute reality. In doing so, we may fool both ourselves and the “malfunctioning” other with regard to what we are doing: We are trying to implement our values without admitting, and perhaps, without even knowing, that this is what we are doing.

(p. 12).

Through a narrative lens, the perspectives of therapists are immersed in the cultural influence

and ideologies that are socially constructed around all of us, thus, their interpretations are as

biased and subjective as those of their clients (Smith, 1997, p. 7). In Philip Cushman’s (1995)

critique of traditional psychotherapy, he says, “when we study a particular illness, we are also

studying the conditions that shape and defined that illness, and the sociopolitical impact of those

who are responsible for healing it” (p.7). Where a traditionally oriented therapist might ask,

“what kind of intervention would you use with [diagnostic category] clients?” and then fit

interventions accordingly, a narrative therapist would not make such a scientific reductionist

statement about the client’s illness or health. This assumption of power and expert stance can

serve to privilege the therapist’s voice and influence the shape of therapy, even diminishing the

client’s voice and creating a dependence on knowledge and support of the therapist and of the

system they are a part of (Smith, 1997, p. 8-16). Smith (1997) describes the contrasting narrative

stance in more detail:

Narrative ways of working can also involve helping clients separate from totalizing, pathological descriptions, or inviting them to explore potentially empowering alternative voices and stories previously de-emphasized in their lives. Thus narratively oriented therapists see their contribution as facilitating a safe, exploratory therapeutic environment where diverse, nonpathological, alternative perspectives and stories can be entertained, rather than as expertly providing clients with authoritative, complete, or definitive responses (p. 4).

The narrative therapist often takes the stance of “puzzling together.” This way of working is

intended to encourage the client’s own understanding and meaning-making to emerge. This

stance is also intended to assist clients in generating more useful and empowering life stories

44

than the problem narratives they often come into therapy with. In contrast to traditional and

systemic therapies which Smith (1997) says, “generally assume that therapists’ objectivity

provides the foundation for their being experts in defining clients’ problems and solutions” (p.

3), narrative therapy positions the client’s problem apart from his/her identity, decreasing self-

blame, de-emphasizing pathology, and empowering the client to take responsibility for and cope

successfully with these problems (Smith, 1997, p. 35).

Narrative approaches to working with children who have experienced abuse address

power and patriarchy within the culture as well as within the life of the child (Adams-Westcott &

Dobbins, 1997). Adams-Westcott and Dobbins (1997) explain how the narrative approach

represents a shift in thinking from the family systems therapy perspective, “instead of locating

the problem inside people or in relationships between people, we began to locate the problem in

restraining beliefs, patterns of interaction, and cultural expectations and practices that create

vulnerability to abuse” (p. 195-196). When people talk about abuse, they talk about it from the

cultural perspectives available to them. The struggle of sexually abused children to make sense

of their abuse is culturally embedded (Mossige, Jensen, Gulbrandsen, Reichelt and Tjersland,

2005). Blending a trauma theory perspective and a narrative perspective Mossige et al. (2005)

say that:

Within both trauma and narrative perspectives, understanding the sociocultural context in which the abuse takes place is therefore imperative. In order to understand the difficulties children face, a narrative perspective needs to include the emotional significance of the events to be narrated. The trauma perspective encompasses the emotional aspects of an event but overlooks the specific cultural impact of the events and previous lack of narration. A theory that intends to understand children’s narration difficulties should encompass both of these perspectives

(p. 401).

45

Thus a narrative approach to trauma work considers the person within their environment,

working with families and individuals to establish narratives about the context within which the

abuse took place.

Mossige, et. al. (2005) studied the way that children narrate the abuse they have

experienced, finding that children describe events that are important to them in detail,

elaborating on the context and their personal experience of the story. In contrast, they found that

children’s descriptions of abuse and other stressful events are often much thinner, shorter, and

involve less plot development. The researchers understood this reality to be a function of the

great contrast between the child’s expectations of the world and the reality of their abuse

experience. In their study of how children’s abuse stories situate them as moral agents within

their culture, Walton and Bruner (2002) point to the importance of the thoughts and knowledge

of the protagonist within a story. Often during abuse, children are told to keep the story of the

abuse hidden and are left only with the ideas and messages of those that have abused them. This

reality narrows their ability to make personal moral evaluations of abusive events and to measure

them against their expectations of the world. Mossige et. al (2005) say, “when the abusive event

is something that ‘just happens,’ the children are left alone with the project of constructing

meaning and with the evaluation of their own moral stance relative to the events. The odds of

them reaching an explanation of how the abuse started and why it happened are very small” (p.

398).

Hence, the narrative framework for telling and retelling stories may be particularly

helpful for approaching clinical work with individuals who have experienced trauma; a way for

these individuals to thicken the narrative of the protagonist that was not developed during the

experience of abuse. Adams-Westcott & Dobbins (1997) say that,

46

The way in which a young person interprets the experience of abuse can have a profound impact on her story about herself. Children and adolescents who have experienced sexual assault are at risk for developing negative stories about themselves that are dominated by the experience of abuse and its effects

(p.197)

Through discussing events with others, children are able to organize events in more coherent,

temporally extended and integrated ways (Fivush, 1998; Nelson 1993). This kind of narration

helps a child to interpret events (Burner, 199l; Fivush & Haden, 1997). From a trauma

perspective, one would not necessarily expect a child to narrate their experiences, for events that

remind children of traumatic experiences often evoke aversive emotional reactions associated

with the trauma. Children may avoid narrating events that evoke these feelings (Koverola & Foy,

1993). Furthermore, a narrative approach does not assume that all children who experience

sexual assault begin to interpret their lived experiences through an abuse-dominated lens

(Durrant & Kowalski, 1990). However, when given the opportunity to thicken their personal

narrative of the protagonist of their story, children may be able to gain a sense of agency that was

stripped form them during the time of abuse.

Traditional psychiatric literature as well as popular literature about childhood sexual

abuse has the potential to reinforce a negative story about the self. Adams-Westcott & Dobbins

(1997) say that, “these models assume that children who experience abuse inevitably suffer long-

term emotional and interpersonal difficulties, and prescribe a specific set of steps that are

necessary to help the children recover” (p. 1999). If adults in a child’s life begin to understand

the abuse in this way, looking for problems within the child who has been abused, they

inadvertently invite these problems to develop. Overtime the children’s actions may be

interpreted as psychopathology and, in turn, the young person may begin to pathologize herself

(Adams-Westcott & Dobbins, 1997). Hence, the clinical approach to individuals who have

47

experienced trauma - the way that the “problem” is located within the therapy - has implications

for treatment outcomes as well as implications for the stories the individual will adopt about

themselves and the world around them. A narrative therapist would carefully listen closely to the

problem narrative and be attentive to bringing in other stories that speak to life before the trauma

occurred. They would position clients as experts in their own lives and assume that they have the

skills and competencies to construct more positive stories about themselves (Goldenberg &

Goldenberg, 2013, p. 396).

For instance, narrative therapists position clients as members of networks and

communities and are curious about the role of other people in a person’s struggle with a problem

(McLeod, 2006). White (2004) has suggested that the narrative therapy is built around “folk

psychology - “a particular tradition of understanding life and identity that is deeply historical” -

rather than being based in contemporary psychological science (p.19). In other words, the

experiences of individuals are understood as complexly brought to the fore by their experiences

within their families, communities, cultures, time in history, as well as other relevant contextual

factors within the environment. Along these lines, Roberts (1994) encourages therapists to pay

attention to family stories indicating that they, “offer possibilities both to interpret and to make

meaning of individual lives and family dynamics, while at the same time observing influence of

historical occurrences on their lives” (p. 1). The factors that would be considered as contributing

to the ecology and etiology of problems through the lens of traditional individual and systemic

therapies would be broadened and textured in narrative therapy; inviting the telling of familial

and cultural stories rather than the telling of the environmental context wherein the problem

takes place. A narrative therapist is interested in the story of the whole person rather than just the

story of the trauma.

48

Externalizing Conversations

Externalizing conversations are a tool used in narrative therapy to support clients in

resolving the problems they have come to therapy with. White (2007) explained that, “many

people who seek therapy believe that the problems of their lives are a reflection of their own

identity, or the identity of others, or a reflection of the identity of their relationships” (p. 9).

White and Epston (1990) created the idea of externalizing conversations, believing that when

people see their problems as a part of themselves, this impacts a person’s ability to solve his/her

problems and can even inflate and intensify them. In viewing problems as reflections of certain

“truths” about their nature or character White (2007) said that, “people come to believe that their

problems are internal to their self or the selves of other – that they or others are in fact, the

problem.” (p. 9). This belief, White said, buries them further into the very problems they are

trying to resolve. Instead, narrative therapy seeks to emphasize that, “the problem is the problem,

the person is not the problem” (Freeman, Epston & Lobovitz, 1997, p. 8).

Through the linguistic practice of externalization, narrative therapy works to separate

persons from problems. Freeman, Epston and Lobovits (1997) indicate that, “separating the

problem from the person in an externalizing conversation relieves the pressure of blame and

defensiveness. Instead of being defined as inherently being a problem, a young person can now

have a relationship with and externalized problem” (p. xv). This process serves to de-pathologize

the person, expanding beyond the limits of interpretation based on objective observation of an

individuals psyche to considering the possibilities that exist within the relationship between the

person and problems. Roth and Epston (1996) explain this further:

In contrast to the common cultural and professional practice of identifying the person as the problem or the problem as within the person, this work depicts the problem as external to the person. It does so not in the conviction that the problem

49

is objectively separate, but as a linguistic counter-practice that makes more freeing constructions available

(p.5).

Take, for example, the stories that Peter told about himself when he gained the ability to see.

Peter, a twelve-year-old boy, had been nearly blind since birth, only able to see a murky picture

of the world around him until he had a series of operations that gave his eyes normal vision for

the first time. A year later, Peter’s parents brought him to therapy because their previously well-

adjusted son was refusing to go to school, having realized what an ugly place the world can be.

Before Peter gained his ability to see the world clearly, he could only see the one or two people

that were close by. Now he was able to see the foreground he knew as well as the background of

his classroom that he had never seen before. This background was full of people laughing,

gossiping, fighting and grimacing; an environment quite different than the one that he pictured in

his mind. Peter was disillusioned and overwhelmed by the new reality he encountered. He

withdrew from his old friends, hating what he saw and wanting to go back to his more familiar

world. By the time the family came in for counseling, they had tried everything they could think

of and were feeling defeated (Monk, 1997).

Their new therapist asked them a few questions which served to externalize the problem

such as, “how has ‘being able to see’ changed things?” and “has sight got in the way of your

friendships?” and “is sight keeping you a prisoner in your home, because before it sounded like

you were free to come and go pretty much as you wanted?” These questions opened the family

up to their understanding of how Peter’s ability to see had impacted their family; how their

expectations of what it might be like for Peter to gain normal vision were not realized by his

ability to see. The family had looked forward to Peter gaining the ability to see with excitement,

hope, and promise. These feeling had motivated their decision to have the operations. By

50

externalizing the problem of “sight,” Peter and his family could now begin to talk about it at a

distance, to see how their expectations of “sight” and their experience of “sight” have impacted

their experience of “sight” as they have come to know it. Given this externalization, they could

begin to build a fuller picture of the plot developments that led to Peter’s operations and their

current frustration. The family could talk about their son Peter and his strengths apart from his

“sight.” For Peter’s family, this was an empowering vantage point from which to observe the

problem (Monk, 1997).

The externalizing process begins within the first session a narrative therapist has with an

adult, child or family as the therapist tries to get to know more about the child. Through inviting

descriptions of the person that exclude the problem, the therapist invites the whole person into

the therapy context, expressing interest in the person’s unique qualities and ideas rather than

simply the problem he/she comes in with. By way of this inquiry, Freeman, Epston and Lobovits

(1997) suggest that the therapist learns about strengths that might be hidden behind the

presenting problem and indicate that, “such discoveries can become the foundation upon which

an alternative story is built, one based upon the child’s and family’s competencies and

sufficiently compelling to stand up against the problem-dominated story” (p. 35). The goal of

externalizing conversations, in part, is to allow the client to speak in his own voice and to work

on the problem himself (Monk, 1997, p. 42). For those that come into counseling with the

expectation that they will have to come to face their inner deficiencies, pain and deficits, this

process is a pleasant surprise. Monk (1997) says that the process of externalization can be

relieving; clients are “immediately given the status of agents in their own and others’ lives and

regarded as resourceful, intelligent persons engaged in common human struggles” (p. 45). The

51

alternative stories created by way of externalizing conversations are built-up and “thickened”

through the narrative process of re-authoring conversations.

Re-authoring Conversations

In concert with their ideas about externalizing conversations, White and Epston (1990)

originated the idea of having “re-authoring” conversations with clients. They noticed that as

therapists and clients co-constructed externalized stories about problems, the client would often

feel able to let go of “chronic, habitual, and disempowering ways of thinking and being” (Smith,

1997, p. 35). Through this process, they found that clients were able to consider new and

empowering self-narratives that had been lying in the background of their dominant problem

narrative, bringing these neglected and unnoticed stories to life. Goldenberg and Goldenberg

(2012) explain that narrative therapists:

Argue that to search for underlying traits or needs or personality attributes is to rely on artificially ‘thin’ descriptions (e.g., superficial, insubstantial descriptors of internal states such as normal/abnormal or functional/dysfunctional) when we should be looking for ‘thick’ (enriched, intentioned, multistoried) descriptions, shaped in part by personal, historical, political and cultural forces

(p. 397).

These ideas of “thin” and “thick” descriptions are central to the narrative process of re-authoring

conversations.

Thin descriptions, in narrative therapy, are understood as descriptions about the self that

are imposed by others with definitional power (teachers, parents, doctors, clergy) and are

integrated into a person’s conceptualization of himself or herself (Goldenberg & Goldenberg,

2012). These descriptions are often superficial, disempowering, and come along with labels such

as lazy, depressed, anxious, selfish, greedy, or crazy. These thin descriptions often translate into

the person making thin conclusions about themselves, such as “I am lazy,” which are problem-

52

saturated and obscure the strengths and positive characteristics about the person (Goldenberg &

Goldenberg, 2012).

In contrast, thick descriptions are stories that are deep, rich, and diverse. They speak to

the unique nature of a person’s history and identity (Goldenberg & Goldenberg, 2012). They are

comprehensive understandings drawn from preferred stories of the person or persons who have

come for therapy. Freedman and Combs (1996) describe the narrative therapist’s effort to thicken

descriptions in the following way:

Narrative therapists are interested in working with people to bring forth and thicken stories that do not support or sustain problems. As people begin to inhabit and live out alternative stories, the results are beyond solving problems. Within the new stories, people live out their new self images, new possibilities for relationships, and new futures.

(p. 16)

Through the re-authoring process of collaboration co-creation of thick descriptions and new

narratives about a person’s life, narrative therapists aim to illuminate the many possibilities

available within the stories people tell about themselves, each other and the world (Goldenberg

& Goldenberg, 2012).

Returning to the story of Peter, the thin description of Peter’s “problem” might have

been, “Peter has had problems at school ever since his sight restoring operations.” This

description brings the symptoms that have developed since Peter’s surgery into light and

amplifies the problems that sight has created in the lives of Peter and his family. In contrast, a

thick, re-authored description of this problem would invite Peter and his family to tell stories of

life before the operations, their hopes for the operations, the experience of the operation itself,

their individual experiences of life since the operation and the role of Peter’s sight as well as his

lack of sight in their family story. Making space for subtle shifts in Peter’s relationship to the

problem would open the door for new understandings and a bigger picture of the meaning the

53

family is making of the changes within their lives. This process, if pursued with curiosity and

persistence by Peter’s therapist, would de-emphasize the “problem” while constructing a history

of Peter’s preferred story, beginning the process of self-redefinition. Through this process, Peter

can replace his stance as “helpless” and gain the stance of “expert” with regards to the story of

his life. In turn, Peter’s therapist will be an important consultant, co-authoring this story and

Peter’s parents will be an important audience to this story. With the support of these narrative

techniques of re-authoring, Peter might leave therapy with more skills for approaching situations

of fear in the future, given his new stance of “expert” in his own life. (Monk, 1997).

Who Can Benefit from Narrative Therapy?

Narrative therapy prides itself on working from a place of de-pathologizing the problems

people bring to therapy and in working with them on externalizing their problem and re-

authoring new stories about themselves. Hence, it was developed for the purpose of diverse

individuals and communities. Critics of narrative therapy have wondered about how narrative

therapy can work with individuals who are mandated to therapy, are significantly compromised

in their language abilities, or who are not linguistically eloquent (Dulwich Center, 2013). The

Dulwich Center, where narrative therapy originated, speaks to these questions and concerns

about the practice on their website:

Narrative therapy/community work always involves conveying meaning and the telling of stories, but the ways in which this occurs differ enormously depending upon the people involved. Much of the work that is now referred to as 'narrative approaches' originated in, and continues to involve work with, very young children. Much of the work also had its origins in conversations with people who had great restrictions upon their lives and ways of expressing themselves (for example, those living within institutions). There is a great diversity of ways in which stories can be told and conveyed that do not require what is generally considered to be eloquence or literacy, or for that matter any formal education. People try to make themselves understood in a great variety of ways. It is the practitioner’s role to engage with the experience and meaning of the person who

54

is consulting them in whichever way or shape the expressions of this meaning occurs

As with any therapy, there are limits within the scope of cross-cultural work as well as with

clients who are mandated to therapy. Walsh (2012) says that “narrative therapy is guided by a

few basic assumptions: that people have good intentions and neither want nor need problems;

and that they can develop empowering stories when separated from their problems and

constraining cultural beliefs” (p. 37). Ideally, this framework, which empowers people to take

active charge of their lives, translates across cultures, communication styles, and personal or

cultural systems of belief.

Conclusions and Key Points

This chapter has reviewed the history of narrative therapy, with special attention to the

roots that narrative therapy has taken in postmodern theory and the framework it has for

maintaining a critical eye towards society. This chapter has also expanded upon narrative therapy

perspectives of diagnosis and pathology, reframing these traditional therapeutic tools and

suggesting a therapeutic perspective with draws problems out of a place within the person’s

identity (externalizing conversations) and encourages the person to create new, unique, and

diverse stories about themselves (re-authoring). It has been exemplified through this chapter that

narrative therapy is not simply a theoretical orientation, but rather a way of practice and being

with people that focuses on the whole person and on bringing the client and therapist together to

co-create the therapeutic space. It is with this framework that the case example in this study’s

discussion will be deconstructed and approached through a narrative therapy lens.

55

CHAPTER V

WINNICOTT’S OBJECT RELATIONS THEORY

“It is in playing and only in playing that the individual child or adult is able to be creative and to use the whole personality, and it

is only in being creative that the individual discovers the self.”

– Donald Woods Winnicott (Winnicott, 1971)

This chapter will introduce the reader to the perspective of object relations theory,

specifically the work of Donald W. Winnicott. The chapter will provide background on

Winnicott and object relations theory and present three concepts which were developed by

Winnicott: the concept of the good-enough mother, the concept of transitional objects and the

concept of the capacity to be alone. Through introducing these particular concepts, the reader

will be primed to consider how Winnicott might think about the phenomenon of imaginary

friends in older children who have experienced trauma.

History

Donald W. Winnicott (1896-1971) was a British pediatrician and a member of the British

School of Object Relations, a school made up of psychoanalysts including Melanie Klein,

W.R.D. Fairbarin, Harry Guntrip, John Bowlby, and others (Mitchell & Black, 1995). As a

pediatrician, Winnicott worked extensively with children, showing special interest in those who

had experienced trauma and other emotional difficulties. From there, he became interested in

psychoanalysis, training under Melanie Klein until he began practicing analysis independently.

A key component of object relations theories, to which Winnicott contributed greatly, is a

belief that individuals possess an internal world of relationships, developed through interactions

56

early in their lives, that may be more powerful than one’s relationships with actual people

(Flanagan, 2008). When talking about individuals, object relations theorists talk both about the

relationships people have in their lives as well as their internal representations of those

relationships, calling both objects, internal and external. These theorists believed that people take

in psychological material just as they physically process sustenance. This psychological material

is then metabolized and ingested in unique ways for each person. The result is internal

representations of object experiences, meaning, the way that people react to experiences

encountered in external relationships in everyday life is influenced by the information they have

internalized, early in their life, about relationships (Flanagan, 2008).

With the foundation of his education in object relations as well as his experiences

evaluating many children with trauma histories, severe emotional difficulties and deprivation,

Winnicott began to evaluate the relationship between infants and their environments (Goldstein,

2001). In keeping with the tradition of object relations of which he studied, Winnicott (1956)

believed that, “drives and impulses could be gratified without the relationship being all that

important, whereas needs have to be met by a person, thereby placing the relationship at the

center of the experience” (Flanagan, 2008, p. 124). His work focused principally on the

relationships between infants and their primary caregivers, very often their mothers. He viewed

these relationships as the basis of emotional development and coined the terms holding

environment and good enough mother to describe the way that he believed a mother must

provide care for her infant (Goldstein, 2001).

Winnicott believe that in infancy, the caregiver is experienced as a ‘bundle of

projections’ and not a real or separate being (Winnicott, 1969, p. 712). He emphasized that a

nurturing environment and supportive environment is necessary for an infant’s optimal

57

emotional development but that this mother need not be perfect, just “good-enough” for the

infant’s emotional capacities to grow in a healthful way. For Winnicott, the quality of the

infant’s early caregiving environment was crucial for the infant’s development of “personhood.”

Representing a large shift from earlier theories, Winnicott believed that psychopathology often

stemmed from failures within the caregiving environment. He developed the term “false self

psychopathology” a process which Mitchell and Black (1995) illustrate describing “the child

shapes a false self that both deals with an external world that must be watched and negotiated

and also shelters the seeds of more deeply genuine experience until a more suitable environment

is found” (p. 133). Winnicott was most interested in clients with false self presentation and

viewed them as having experienced terrible gaps in their experience of holding and mothering.

He described that these clients would often go through the motions of acting like a person yet not

feel to himself like a person, lacking personal meaning and identity. (Mitchell & Black, 1995).

In a radical shift from the work that had been done so far in history, and in a way that was

too provocative for many thinkers of his time, Winnicott proposed definitions for styles of

parenting which could promote or inhibit optimal emotional development and mental health

(Goldstein, 2001). In addition, he did not envision development as a predictable sequence of

stages that build on and replace the stage before, but rather a process of continual regeneration.

He described a continuum of development between subjective omnipotence and objective reality

with a third form, called “transitional experience”, lying between (Mitchell & Black, 1995, p.

127). Mitchell and Black (1995) describe subjective omnipotence as a form of being in

relationship through which “the child feels she has created the desired object, such as the breast,

and believes she has total control over it” (p. 127). On the other side of the continuum, Winnicott

believed that in objective reality, the child looks for desired objects out in the world and is aware

58

of his/her lack of control over desired objects and his/her ability to find them. The transitional

space is somewhere in between where the child is aware that he is not omnipotent yet is not fully

able to negotiate his desires by way of other people (Mitchell & Black, 1995). This transitional

experience will be explored further later in this chapter.

Overall for Winnicott, optimum development was made possible by the ability of a

caregiver to meet the needs of the child including the need to be seen as valued and unique,

whole with good and bad aspects, cared for, protected and loved (Flanagan, 2008).

The concepts of the good enough mother and holding environment

The concept of “The Good Enough Mother” was born out of Winnicott’s belief that, at

the very beginning of life, infants thrive with a mother who can allow herself to intuit the child’s

needs and desires and shape the world around the child so as to fulfill those desires; to “bring the

world to her infant” (Mitchell & Black, 1995, p. 125-126). He believed that a healthy mother

must lose herself completely in her baby in a process that he called “primary maternal

occupation”. Through the biological process of pregnancy and childbirth, the mother becomes

conditioned to intuit and attend to the baby’s every need, to lose herself in the needs and

subjectivity of the infant. It is through this attunement experience that the baby is able to develop

an internal world of subjectivity that allows her to experience human life as real and meaningful

(Mitchell & Black, 1995).

According to Winnicott, a mother did not need to be perfect at this, just “good enough”

for her infant to know that he/she is safe in the union of her love (Flanagan, 2008). The process

required a good enough holding environment which, for Winnicott, did not mean only the literal

holding a mother does of her child, “but the capacity of the mother to create the world in such a

59

way for the baby that she feels held, safe, and protected from dangers without and protected as

well from the dangers of emotions within” (Flanagan, 2008, p.130-131).

In order to provide such a holding environment, Winnicott believed that parents needed

to develop a functional attunement with their infants; to be aware of their infants needs and to

exercise the capacity to meet those needs. In describing Winnicott’s ideas, Mitchell and Black

(1995) say, “as the infant’s needs and wishes emerge from the unintegrated drift of

consciousness, the good-enough mother intuits the child’s desire relatively quickly and shapes

the world around the child so as to fulfill that desire” (p. 126). Within this holding environment,

the infant needs his or her caregiver to be reliable and consistent. Despite inevitable failures, the

caregiver must keep from excessively intruding upon or depriving the infant. This good enough

mothering, Winnicott emphasized, results in healthy psychological development for the infant. In

contrast, maternal failures are cause for ego defects and the creation of a false self that arises in

an effort to please others (Goldstein, 2001). Winnicott believed that the development of a true

self, the core of a person, can only thrive in the presence of maternal attachment that nurtures

and allows the child’s individuality to grow (Flanagan, 2008).

According to Winnicott, emotional development is compromised when an infant’s

caregiver is unable to provide a good-enough holding environment. Instead of feeling held within

a protective space to play and grow, the child has to prematurely deal with the outside world in a

process Winnicott called impingement (Mitchell & Black, 1995, p. 129). If this maternal failure

is chronic, it can result in a split within the internal world of the infant through which they

develop a disconnect between their true self and their false self. In this way, Mitchell and Black

(1995) describe that the not-good-enough mother provides the child with, “a world he has to

60

immediately come to terms with, to adapt to, and the premature concern with the external world

cramps and impedes the development and consolidation of the child’s own subjectivity” (p. 129)

In treatment, Winnicott viewed the therapist as an agent to heal this split, providing a

holding environment for the client much like the good-enough mother, creating a space in which

the therapist can deeply understand and attune to the client’s experience. Within this space, the

client could be given the proper space in which to “be” and to experience a nurturing and safe

holding environment where the true self could begin to emerge. In treatment, the therapist, like

the good-enough mother, “tries to grasp the deeply personal dimensions of the patient’s

experience, the patient’s spontaneously arising desires” (Mitchell & Black, 1995, p. 133). She

provides an environment where her own needs and subjectivity are on hold and the arrested

development that began in infancy can begin to grow again. The therapist works to provide the

patient with a space just to “be” and connect and express their experience. Through treatment,

the person is provided a context through which their true self can begin to reemerge. For

Winnicott, treatment is possible because of the therapeutic relationship, which naturally

facilitates a regression that awakens old developmental needs that were missing in childhood.

Through this relationship, the client has a new experience of relating, and he or she actively

molds the treatment to provide that which was missing. The result is a revitalization of the self

(Mitchell & Black, 1995).

The concept of transitional objects

A crucial stage of development in infancy, according to Winnicott, is the transition from

total dependence on the mother to increased independence and the ability to negotiate the

intermediate stages there within. For Winnicott, this was a relational shift from total dependence

as well as a cognitive shift from “subjective omnipotence” to greater independence and an

61

“objective reality,” allowing the infant to both attach in early attachment relationships as well as

separate from them maturely (Mitchell & Black, p. 127). Winnicott believed that good-enough

parenting enabled infants to internalize the soothing functions of their mother, “as long as the

mother’s unconscious wishes do not interfere with the baby’s need to merge and then to

differentiate, the process can go on” (Tolman & Lane, 2007, p. 41). Good enough mothering,

“lays the groundwork for the integration of the mother object in each person’s inner world, a

piece of that psychological apparatus that eventually allows one, through mental functioning to

perform these care-giving activities for one’s self” (Tolman & Lane, p. 41). In this way, the

infant can learn to self-soothe when the mother is no longer available (Tolman & Lane, 2007).

Winnicott believed that transitional objects, such as a blanket or stuffed animal that is

associated with the mother, could stand in as an emotional bridge between infants and mothers

during this transitional time of development (Goldstein, 2001). Because this separation-

individuation process is such an anxiety provoking experience, Winnicott believed that a child

takes on an object, representing the good-enough mother, in order to self-soothe while

maintaining their growing independence. Winnicott’s theory asserted that the transitional object

“constitutes a special extension of the child’s self, halfway between the mother that the child

creates in subjective omnipotence and the mother that the child finds operation on her own

behalf in the objective world” (Mitchell & Black, 1995, p. 128).

Winnicott talked about transitional objects as, “literal objects that a child carries with

them in order to begin to cross the great gap from complete union with their caregiver toward a

sense of self as a separate entity” (Flanagan, 2008, p. 131). He understood these objects as

physical and described the qualities that all transitional objects are required to contain:

the infant assumes rights over the object… the object is affectionately cuddled as well as excitedly loved and mutilated… it must survive instinctual loving and also

62

hating… it must seem to the infant to give warmth… in health the transitional object does not go inside… it loses meaning, and this is because the transitional phenomena have become diffused (Winnicott, 1971, p. 5).

Winnicott believed that the object becomes especially useful for the infant at bedtimes or at

times “when a depressive mood threatens” (p. 4), allowing the child to experience the illusion of

their caregiver’s presence while the caregiver is not actually available. Foehrenbach, Celantano,

and Lane (1997) say that, “normally, the infant substitutes security ‘blankets,’ and in doing so

chooses an object that embodies the essence of the mother-her smell, her touch- but is in reality a

symbol of her soothing and protective function” (p. 43). As the child matures, the type of object

often evolves to more structurally represent the caregiver, such as a doll or stuffed animal.

Finally, the development of language, and the term “mommy” is argued to be a final transitional

object between the separating infant and mother, allowing for the internalization of the mother-

object in a way that can be carried permanently (Foehrenbach, Celantano, Kirby, & Lane, 1997).

The development of a transitional object improves the infant’s ability to be alone. Due to

the representation of the caregiver, the transitional object holds qualities of the caregiver and can

provide the comfort of the caregiver even when the infant is alone. The comfort they derive from

this object bolsters their sense of their capacity to tolerate difficult and emotional situations

effectively on their own. If, on the other hand, the infant has been the victim of too much

aloneness, isolation or maladaptive parenting, their inner world can be filled with emptiness and

fear (Flanagan, 2008).

Winnicott (1971) writes about a 7-year-old boy whom he saw in therapy that was

preoccupied with string. The patient used string to tie object together, to threaten his own life and

to threaten the life of his sister. His parents recognized this preoccupation, but could not

understand its significance. Through treatment, Winnicott helped the family to realize that the

child thought about the string most when he felt afraid that his mother would leave him. When

63

the patient was just over three years old, his younger sister was born; this was the first rift in his

experience of closeness with his mother. Later on in that year, his mother was hospitalized for an

operation and then again the next year she was hospitalized for two months for depression.

Through testing, Winnicott found that that boy appeared preoccupied with string, formulating

this preoccupation as having to do with fear of separation. Winnicott encouraged the mother to

talk to her son about his feelings around separation when the time presented itself. When she did,

the string preoccupation reportedly ceased. Winnicott (1971) writes of the mother,

she has had many other conversations with the boy about his feeling of separateness from her, and she made the very significant comment that she felt the most important separation to have been his loss of her when she was seriously depressed; it was not just her going away, she said, but her lack of contact with him because of her complete preoccupation with other matters

(p. 12).

Each time the mother needed to separate form the boy for a hospital stay or due to a depressive

episode, the boy’s string play began again. Each time, the parents were able to have a

conversation with the boy about his feelings of separation and the string playing ceased.

As the child grew older, he began to tend to multiple stuffed animals as if he were their

mother. He did this more so when his own mother was away. Winnicott commented on concern

that the child would not move beyond this transitional experience and that his preoccupation with

string and his animals could turn into a perversion (Winnicott, 1971, p. 13). He later commented

that the child’s separation issues never did resolve and the child could not leave home without

running back. Later in life, he developed new addictions, especially to drugs, and could not leave

home to receive an education. Winnicott linked this back to an inability to separate and

individuate from his mother and a deeply seeded connection to the cycle of her depressive

episodes. The child, and later the adolescent, used the string to cope with intolerable feelings of

fear when in the absence of his mother; a self-soothing object that supported the child in feeling

64

tied to his mother. Winnicott found this case to be beyond treatment, finding that the treatment

did not intervene early enough and the child was unable to develop object constancy and self-

sooth in the absence of his mother (Winnicott, 1971).

Later in the same paper, Winnicott described his treatment with an adult woman who,

through the course of therapy, recalled images she had as a young girl. The client had

experienced a traumatic separation from her parents at age eleven and began developing these

images thereafter. Winnicott said that these images could easily be labeled hallucinations, except

for her age. The woman described seeing an angle by her bedside, an eagle chained to her wrist,

and a white horse. Winnioctt (1971) writes of the patient

She also had a white horse which was as real as possible and she ‘would ride it everywhere and hitch it to a tree and all that sort of thing’. She would like really to own a white horse now so as to be able to deal with the reality of this white horse experience and make it real in another way.

(p. 16)

Of the formulation of this case Winnioctt (1971) wrote,

Here was the picture of a child and the child had transitional objects, and there were transitional phenomena that were evident, and all of these were symbolical of something and were real for the child; but gradually, or perhaps frequently for a little while, she had to doubt the reality of the thing that they were symbolizing. That is to say, if they were symbolical of her mother’s devotion and reliability they remained real in themselves but what they stood for was not real. The mother’s devotion and reliability were unreal.

(p. 17).

Winnicott and the client understood her animal images as representing the good parents that she

had and lost while the chain on the eagle represented the woman’s fear of loosing that good and

comforting objects of her parents as represented by the animals (Winnicott, 1971). This example

of imagined transitional phenomenon is unique in the literature, a departure from Winnicott’s

65

description of transitional objects as only physical; in this case, the transitional phenomena is

imagined.

With regards to development Winnicott elaborated on the experience of the child who

cannot flow through the different forms of the developmental continuum. He described a person

who lived primarily in subjective omnipotence as autistic and self-absorbed, lacking the ability to

bridge their experience of the world with objective reality, such as the seven year old boy who

could not make use of a comforting sense of his mother in her absence. On the other end of the

continuum, a person who lived in objective reality was artificially adjusted and lacked passion

and originality such as a child who cannot engage in imaginative or creative play. He believed

that the transitional experience was, as Mitchell and Black (1995) describe, “a protected realm

within which the creative self could operate and play; it was the area of experience where art and

culture were generated” (p. 128). This realm, for Winnicott, was crucial in the development of a

deeply rooted self that could connect creatively with the world.

The concept of aloneness

Winnicott (1958) believed that the capacity for an individual to be alone is one of the

most “important signs of maturity in development” (p. 29). He distinguished this capacity to be

alone from the act of spending time alone, believing that one can be alone and suffer alone or be

alone and learn to enjoy the solitude. Winnicott (1958) traced the capacity to be alone to early

childhood when, “the infant is able to do the equivalent of what in and adult world would be

called relaxing” (p. 34). He found it paradoxical but true that in the presence of another person,

the infant is then, “able to become unintegrated, to flounder, to be in a state where there is no

orientation, to be able to exist for a time without being either a reactor to an external

impingement or an active person with a direction or interest in their movement” (Winnicott,

66

1958, p. 34). He believed that this foundational ability to be alone is the ground on which

emotionally maturity and sophisticated aloneness are built.

Winnicott (1958) said that this internal capacity to be alone relies upon the existence of

an internalized good object and that

the relationship of the individual to his or her internal objects, along with confidence in regard to internal relationships, provides of itself a sufficiency of living, so that temporarily he or she is able to rest contented even in the absence of external objects and stimuli

(p. 32).

Winnicott went on to say that this capacity is reliant on an experience of good-enough mothering

and satisfaction of instinctual gratifications. If the individual experiences too much persecutory

anxiety, this process is not possible. The good internal objects must be in the individual’s

internal world and must be available for projection when they are needed. Hence, the capacity to

be alone has its roots in the experience of being alone in the presence of a mother or a mother-

substitute (Winnicott, 1958). Describing how this happens, Winnicott (1958) said,

being alone in the presence of someone can take place at a very early stage, when the ego immaturity is naturally balanced by ego support from the mother. In the course of time the individual introjects the ego-supportive mother and in this way becomes able to be alone without frequent reference to the mother or mothersymbol

(p. 32).

It is through this experience that a person learns to understand that when they are alone, they are

never truly alone because they have internalized the feeling that someone else is there. The

person’s internal environment is regulated by their experience of having been repeatedly

accompanied in infancy, without demands placed on them, by a loving mother or mother-

substitute. Through the experience of the good-enough mother providing for the infant’s

instinctual needs, the good object of the mother is internalized and the person believes that they

67

need not be anxious when alone. When good internal objects are available for projection at a

suitable moment; the person has what they need within them to meet their needs when they’re

alone. “Paradoxically” Winnicott (1958) says, being alone is something that, “always implies

that someone is there” (Winnicott, 1958, p. 34).

Winnicott also wrote about the experience of individuals who did not have good-enough

mothering and did not develop the capacity to be alone. Flannagan (2008) summarized his

thoughts saying, “if aloneness is experienced as too empty, separate, or bleak, it becomes

unbearable” (Flanagan, 2008, p. 132). He talked about the importance of a child’s inner world

being peopled by comforting figures, and the painfulness of when it is crowded by threatening,

controlling figures who offer neither safety, nor peace (Flanagan, 2008).

68

CHAPTER VI

DISCUSSION

“…the companion can be neither completely integrated… nor completely abandoned, for to

abandon him would be to lose an important part of the self, while to integrate him is too

conflictive and beyond the synthetic capacity at this time”

– Bach (1972, p. 170)

Recap of Major Points and Theories

This thesis began by discussing the phenomenon of imaginary friends as they present in

the lives of adolescents with histories of attachment trauma. The phenomenon chapter included a

review of psychoanalytic and psychological literature about the phenomenon. It also reviewed

trauma research, providing scaffolding for understanding how attachment trauma may influence

the presentation of the imaginary companions in adolescents. Finally, the chapter introduced and

summarized the case of Ben, written by Stephen Proskauer, Elizabeth T. Barsh and Lucita B.

Johnson. Ben’s case will be used in the present chapter to ground this author’s discussion.

After introducing the phenomenon, this author introduced two theoretical approaches that

will be used as lenses through which to understand the phenomena, specifically as the

phenomenon presents in the case of Ben. The first theory chapter provided a brief history of

narrative therapy theory, and summarized the concepts of re-authoring and externalizing

conversations; all of which are foundational components of narrative therapy. The second theory

chapter introduced Donald W. Winnicott, a psychiatrist and object relations theorist and a brief

history of the school of object relations and Winnicott’s role there. Three of Winnicott’s

69

concepts were introduced: the good-enough mother, transitional objects and the capacity to be

alone.

This discussion chapter will bring the phenomenon and two theory chapters together

through a discussion of treatment approaches to the phenomenon. Rather than compare and

contrast, this thesis will offer multiple perspectives through which to view the phenomenon as

well as offer ideas to consider for treatment. A significant limitation of this study is that this

author was not Ben’s therapist and the case is not known intimately as it would be if this author

were the identified psychotherapist. Furthermore, Ben was treated by hypnotherapy to which he

appeared to respond to positively. The theoretical approaches offered by narrative therapy and

Winnicott are quite different than the approach of hypnotherapy, thus, we can only postulate the

response Ben might have had to an alternative treatment. Given that this thesis takes on a

retroactive perspective on formulation and treatment, the content will represent ideas and

suggestions for practitioners rather than spend too much time postulating about how Ben might

have responded to alternative treatments. The goal of this discussion is to expand the

conversation about formulation and treatment of the phenomenon by considering

psychotherapeutic perspectives which speak to adolescent responses to attachment trauma,

considering the adolescent’s environment and it’s role in shaping adaptive and maladaptive

responses, and moving away from models that primarily emphasize individual pathology. It is

not within the scope of this research to carry out these ideas with a current case example.

This discussion chapter will begin with an analysis of the phenomenon from narrative

theory and then transition to an analysis of the phenomenon from Winnicott’s object relations

perspective; with attention to the specific concepts highlighted in the theory chapters. For

narrative therapy these are postmodern thoughts, externalizing conversations and re-authoring

70

conversations. For Winnicott, these are good-enough mothering, transitional objects and

aloneness. Aspects of formulation and treatment will be considered within both theories as they

apply specifically to the case of Ben and more globally to the phenomenon. Research and

findings from the theory chapters will be interwoven and additional relevant research and ideas

will be interspersed within the analysis. The discussion chapter will then synthesis the findings

of the analysis, discussing how the application of the chosen concepts within narrative therapy

and Winnicott offer a new way of understanding the phenomenon. Strengths and weaknesses of

the present study will be discussed and areas for further research will be suggested. Finally, the

discussion will consider the implications of these findings for social work practice, policy and

research before offering concluding comments.

Analysis – Examines / Discusses / Connects the phenomenon to the theories

Ben did not speak at first when he presented at the mental health clinic where he was

brought for treatment. The treatment providers tried reassurance and mild medication; however,

these attempts did not result in spoken communication from Ben. His therapist, Stephen

Proskauer decided to hypnotize Ben, at this time Ben’s companion spirit began to speak for him.

Proskauer, Barsh and Johnson (1980) share, “the therapist (S.P.) accepted the companion spirit’s

account without question” (p. 164). With the exception of speaking in the voice of his

companion spirit while under hypnosis, Ben reportedly remained mute until he and his family

did what the companion spirit asked of them. Ben’s initial presentation and response to these

initial interventions is important information to work form when considering how a narrative

therapist or one influenced by Winnicott might begin to work with Ben.

71

Narrative

The theory chapter on narrative therapy introduced the foundational beliefs that were

collated by David White and Michael Epston when constructing the therapeutic approach of

narrative therapy. Bringing these ideas back to the fore, the reader is reminded that, “the

narrative counselor is not concerned with diagnosing ‘the problem’ or with offering ‘treatment.’

The objective is not to find a ‘solution’” (Drewery & Winslade, 1997 p. 41). In narrative therapy,

“much of the skill of the therapist lies in attending closely to the ways we use language: to the

positioning we call people into by the words we use and the ways we organize our sentences” (p.

33). This involves attention to who is speaking as well as the content of their speech. Narrative

therapy is concerned with meaning-making and supporting the client to speak in his/her own

voice. The therapist uses narrative questions to support the client in finding alternative stories

about themselves outside of the story of the problem they came in with. In this way, the narrative

therapist enables the client to work on the problem his/her self (Smith, 1997).

As the reader may remember, narrative therapists believe in the importance of

postmodern thinking wherein the therapist explores the conditions that define and shape a

problem, de-pathologizing the client and inviting them to explore ways of making meaning about

themselves outside of the limited perspective of medicine, science and other narratives with

locate problems within people (Smith, 1997). Within narrative therapy, the role of the therapist is

to, “work collaboratively with people in identifying those ways of speaking about their lives that

contribute to a sense of personal agency, and that contribute to the experience of being an

authority on one’s life” (White, 1995, p. 121). The emphasis on personal agency is particularly

relevant to individuals who have been dominated by others and by problem-saturated accounts of

“self” that have been internalized. When abuse takes place, people often begin to understand

72

themselves by the perspectives of others or from the perspectives of problems in their lives

(Walton & Bruner, 2002). With this refresher of the foundational beliefs of narrative therapy, let

us turn to the case of Ben.

Ben’s parents brought him in for treatment in a state of crisis. He had recently been

through a number of significant transitions (returning from school, moving back in with his

family, cancelling a summer trip) and then his father experienced a “psychotic episode” and

became physically assaultive towards Ben and his mother. On the day he came into the mental

health clinic where he was treated, he had just woken up from a frightful nightmare and was not

communicating through speech. A narrative therapist would encounter Ben’s silence as a form of

communication in and of itself. Perhaps it is an indication of a feeling state such as fear, anger or

sadness or perhaps it is a reaction to the traumatic incidents of the preceding day’s events.

Proskauer, Barsh and Johnson (1980) write, “in the initial interview, he made a drawing of the

black female figure which had threatened to kill him in the dream. Ben labeled the sketch as a

face, but it also resembled a nude female torso with prominent nipples” (p. 163). This drawing

was another form of communication used by Ben, a vehicle with which to share his internal

experience. Without words, Ben has already communicated a story to the therapist; it was a story

of fear, darkness and a lost sense of safety. The reader is reminded to what Allen (2011) writes

about trauma, “the essence of trauma is feeling terrified and alone” (p. 4) and “attachment

trauma damages the safety-regulating system and undermines the traumatized person’s capacity

to use relationships to establish feelings of security” (p. 22). In the night, Ben had experienced a

nightmare when he was all alone, awaking at home to his mother, “hostile” and “controlling” as

Proskauer, Barsh and Johnson (1980) describe her, and to his father who had attempted to

strangle him a short time ago (p. 168).

73

Proskauer, Barsh and Johnson (1980) write that, “Navajo culture has woven into its very

fabric a tendency to dissociative phenomena, exemplified by the nature and influence of

witchcraft beliefs as well as by the prevalence of hysterical seizures” (p. 153). Furthermore, the

authors indicate that Native American cultures recognize and revere the spiritual powers of

natural forces such as those in plant, animals and natural forces. Within the framework of Ben’s

Navajo identity, the idea that he might have been visited by or possessed by a spirit could have

important spiritual significance for Ben and those in his community. The story of Ben’s

experience with a spirit companion would have a different interpretation in a Navajo context,

which emphasize spirituality and natural forces, than it would in a medical community that

focuses on symptoms, pathology, diagnosis and treatment. The reader is reminded that Narrative

therapy is grounded in post modern thinking and the observation that our social world is like a

fabric, tightly woven by threads of belief about what is acceptable and unacceptable, as well as

which ideas and people are relevant and worth following. With that in mind, there were stark

cultural differences between the Navajo cultural upbringing Ben experienced at home and his

experiences at the white man’s boarding school. Ben was likely to have experienced tension

when navigating the cultural contexts he moved between with regards to home and school as

well as an unconscious conflict over integrating or rejecting these multiple identities depending

on the context (Proskauer, Barsh and Johnson, 1980). Keeping in mind these cultural and

environmental factors a narrative therapist would want to engage with Ben and his family

curiously and puzzle with them about the many layers of Ben’s life experiences, helping Ben and

his family to expand the stories they tell from multiple perspectives and encouraging Ben

through questioning and narrative exploration to cultivate an empowered sense of his unique

identity.

74

A narrative therapist would likely take note of the communication inherent in Ben’s silent

presentation on intake and choose to approach Ben’s family to begin treatment, facilitating a

family intervention rather than an intervention solely focused on Ben and his current

presentation. Freeman, Epston and Lobovitz (1997) say, “in narrative therapy, the therapist and

the family may start with an externalizing conversation about the problematic situation before

the therapist tries to get to know more about the child” (p. 35). In this way, the therapist and the

family could deemphasize the problem they came in with as a problem belonging to Ben and

orient the conversation in a broader story. The therapist might ask about the family’s history and

context considering the stressors current impacting the family system, Ben’s history of distress

and coping, and the family’s history of managing problems. Furthermore, the therapist may ask

the family about what Ben is typically like, excluding the problem all together and inviting

descriptions of Ben to get to know him a part from the problem. This approach would give

respect to Ben as a whole person and invite alternative narrative to that of the problem that

brought the family to the clinic. Furthermore, getting to know Ben apart from the problem he

came in with allows the therapeutic system an opportunity to uncover the effect of the problem

on the person (Freeman, Epston and Lovobitz, 1997).

The idea of this early family intervention would be to get a picture of whom Ben is when

the problem is not over taking him. Furthermore, it would actively engage Ben’s parents in his

treatment, a practice that is common and often quite helpful for children dealing with serious

problems. When discussing the problem that brings a family into therapy, Freeman, Epston and

Lobovitz (1997) say, “an externalizing conversation is invaluable here, as the problem can be

referred to in a way that separates it from the young person’s identity but does not ignore it” (p.

36). Furthermore, when Ben did speak, he spoke through his companion spirit, perhaps with an

75

understanding of the dangers Ben would be exposed to if he spoke in his own voice. From a

narrative perspective, by accepting the account of the companion spirit without question, Ben’s

therapist accepted a way of speaking about Ben’s life that gave an explanation for his muteness,

and paved a path toward the development of a new form of personal agency. In this way, Ben

made meaning of his own experience and identified ideas for moving forward vis-à-vis the

mandates of the companion spirit. In essence the companion’s spirit’s request for Ben and his

family to go to Spider Rock was a request to slow the chaos down in Ben’s life and for his

caregivers and girlfriend to focus on accompanying him through his present experience.

Early in treatment, the narrative therapist would want to attend to Ben’s communication

while silent by asking permission of Ben and Ben’s parents to have a conversation about Ben in

his midst, without talking about the problem that brought them to the clinic. In this way, Ben

could be witness narratives about him, outside of the problem narrative and could, perhaps, feel

some accompaniment and holding by his caregivers. Furthermore, this would also allow the

therapist to discover events in the context such as significant loss, losses not attended to in the

family system, major adjustments that might be in process. This inquiry might decentralize Ben

as the one who has the problem. The therapist could ask permission of the family saying

something such as,

“I am hearing that the muteness and nightmare are very worrisome to you, and we’ll certainly keep addressing both of these important experiences. Would it be alright if I took a little time here to find out more about who Ben was before the problem came to have such a profound influence over his way of being… what is Ben like when he is not under the influence of the problem that brought him here today?”

The therapist could also inquire about Ben’s traits and history by saying something such as,

“If we put the problem in a box and ignored it for a while, what would you tell me about Ben that’s important?”

76

In this way, the conversation could open up to an inquiry about Ben’s strengths, activities he

enjoys and aspect of Ben that are appreciated by his family members. Furthermore, Freeman,

Epston and Lobovitz (1997) say, “specific knowledge of [the child’s] interests and abilities tells

what a child might bring to put against the problem” (p. 37). It is the knowing about and giving

voice to these resources and will be particularly helpful in the re-authoring process.

Ben and Ben’s Environment:

We know enough from the history presented in the case study to understand that Ben has

been successful by dominant society’s standards. He has excelled in school, navigated transitions

between home and school environments, managed to be independent of his family at a young age

and, more recently, has found the company of an intimate relationship with a girlfriend. Ben has

made accomplishments, learned skills and achieved at the “white man’s boarding school”. These

successes have been recognized within the boarding school community and he has been

rewarded by good grades and opportunities such as a trip to Europe. However, there is not an

indication of the meaning made of these accomplishments by Ben’s family and Navajo

community (Proskauer, Barsh & Johnson, 1980). Perhaps success in the white man’s world has

strained Ben’s connections with the Navajo world, moving him farther from his roots and

cultural interests. That he did not go to Europe indicates that he was being pulled away from a

very important place to him by these successes.

Narrative therapy stresses that just as there is a complex relationship between a family

and a problem, there is also a complex relationship between the problem and the wider social

forces at play in the person’s life (Rosenwald & Ochberg, 1992). For Ben’s family, boarding

school meant that Ben was away from home for long stretches of time since the age of 7, that he

learned from textbooks and cultivated a knowledge base that would make him successful in the

77

western world. Ben’s time at boarding school meant less time invested in learning Navajo crafts,

skills, spiritual and cultural practices as well as time away from his cultural community of origin

(Proskauer, Barsh & Johnson, 1980). It is important to keep track of the social and cultural

context in which the problem occurs lest we assume that the problem is solely based in the

individual or within the family system (Freeman, Epston and Lobovitz, 1997). For Ben, this

could mean, not only a consideration of his Navajo cultural background and his experience at a

“white man’s boarding school,” but also a consideration of what is expected in both contexts as

far as gender performance, responsibility, leadership, self-expression, emotional health, and

other variants of the expectations placed on Ben. One might wonder what Ben has learned about

emotional expression as well as the expectations of him as a Navajo adolescent male at home,

school, in the community and within the context of grief and loss as Ben lived through a great

deal of loss early in his life.

The narrative therapist working with Ben and his family would need to situate themselves

in the cultural context as well, acknowledging that they are not an unbiased observer in the

therapy or in the creation of Ben’s narrative within therapy. Freeman, Epston and Lobovitz

(1997) say, “as therapists we need to examine our own biases, agendas, and values and make

them available for comment by our clients, colleagues, and ourselves. Our collaborative aim is to

expose the dilemmas that silence and separate us” (p. 56). Furthermore, narrative therapists often

take the stance of “puzzling together” or “co-authoring”; a way of working that encourage the

client’s own understanding and meaning making to emerge (Smith, 1997, p. 4). To open dialogue

such as this, the therapist might want to ask questions that evoke meaning-making for the family,

for instance, “as Ben’s parents, what was it like for you when he woke up this morning feeling

fearful and unable to speak?” or, “can you connect Ben’s current state to anything that has

78

happened recently in Ben’s life or the life of your family?” or “are there stories that you can tell

me about your family that would tell me what’s important to you?” Such questions pull on the

family’s values and the stories they tell about themselves in such a way that positions the

therapist in an important place of “not-knowing”, of curiosity and openness, and of an interest in

stories of the family outside of the problem which is causing them so much worry. Drewery and

Winslade (1997) write, “the counselor is not expected to be an ‘expert’ but rather to be a curious,

interested and very partial participant (rather than claiming an objective neutrality in regard to

the forces that shape the problem) in the production of the meanings of the client’s life” (p.42).

Engaging this deconstruction process, made possible by curiosity, thoughtful questioning and

openness, naturally invites a focus on the people in the room rather than the problem they

brought in with them. Drewery and Winslade (1997) say, “it mobilizes the client’s resources

against the problem” (p. 45). The narrative device of externalization targets the specific process

through encouraging clients to relate to themselves as agents in their own and other’s lives,

encouraging them to regard themselves as resourcefully engaged in the relationships with others

and in the common human struggle (Drewery & Winslade, 1997)

Externalizing:

As a refresher to the reader, Freeman, Epston and Lobovitz (1997) describe externalizing

conversations saying,

When a young person or other members of a family are trying to tell us their story and convey how burdensome a problem is, we might feel the urge to solve or normalize it. However, if we jump too quickly into searching for solutions or try to put a positive face on the situation prematurely, family members are likely to feel that we have not appreciated the depth or quality of their struggle. An externalizing conversation offers a way to listen closely and join with the family, without confirming limited or pathologizing descriptions. The effects of the problem are mapped in detail, along with the family’s preferred way of being and its success in constraining the problem. As we attend to and explore the nuances

79

of their experience of the problem in our dialogue, the effects of a problem-saturated story on family members and relationships becomes evident

(p. 48).

Externalizing conversations are conducted with what narrative therapists call an “externalizing

grammar” (Freeman, Epston & Lobovitz, 1997, p. 57). The idea is to set people’s identities a part

from the problems they come in with and to invite a consideration of the effects of viewing

problems form multiple perspectives. Metaphor is often useful here. For instance, a narrative

therapist might say to Ben and his family, “if we were to give the problem that brought you here

a name, what would we call it?” or “what might we call the issue that brought you here today?”

Given the opportunity to name the problem, the family and therapist can then begin to talk about

it apart from the person. Of course, we do not know what Ben and his family would have

answered to such a question; however, we can imagine that inviting he and his family talk openly

about the issues that brought them in would invite collaboration in understanding and later

addressing the multiple layers which have contributed to the development of those issues

(Freeman, Epston & Lobovitz, 1997).

Given that Ben was not speaking when he arrived at the mental health clinic, one way

that a narrative therapist might seek to engage him in externalization at the outset is through

expressive art. Expressive art therapy is especially helpful, as Freeman, Epston and Lobovitz

(1997) say, for “those who have experienced a constriction of verbal expression at certain times”

and for “children who are not very verbal in therapy” (p. 146). In line to the hypnosis therapy

that was used within the case study, expressive art could support Ben in opening up further

images of his experience. We know that Ben did draw one picture at the outset of treatment;

perhaps this was the best available way that Ben for communicating at the time. Freeman, Epston

and Lobovitz (1997) say that, “the ‘expression’ of problems in art form is inherently akin to the

80

practice of externalization. The very process of drawing, sculpting, or dramatizing the

relationship with a problem naturally evokes a visceral sense of the problem as located for

reflection outside of the self” (p. 147). Furthermore, the act of drawing can be beneficial in itself

as it gives children the opportunity to express their feelings in a physical way; in this way,

artistic expression can offer an immediate release (Freeman, Epston & Lobovitz, 1997). Other

forms of expressive art could also have been helpful to Ben including, the opportunity to draw

the feelings he was experiencing, his family picture, or the parts of himself. Ben’s therapist

would engage curiously with these drawings, asking Ben questions to expand upon the meaning

he might make of what he drew rather than offering interpretation or opinion as might be the

practice in other forms of therapy (Freeman, Epston, Lobovits, 1997). In this way, the therapist

and Ben could take time to build trust and communication, eventually peeling away the layers of

fear and anxiety that were protecting Ben when he came in to the mental health unit.

Re-authoring

Returning to the idea of “puzzling together” and “co-creating” narratives, the narrative

concept of re-authoring is the process by which the therapist and the client work together to

consider new and empowering self-narratives about the client. Narrative therapists believe that

these alternative self-narratives live within the client and are obscured by the dominant problem

narratives that circulate within people’s lives (Goldenberg & Goldenberg, 2012). Within re-

authoring, narrative therapists talk about “thin” and “thick” descriptions. Thin descriptions are

those that are imposed by others with definitional power (teachers, parents, doctors, clergy) and

are integrated by the person into their conception of himself or herself. A thin description is often

a factual account which is blind to the deeper layers of the person it is describing. Within Ben’s

life, a thin description may have been that his father is an aggressive schizophrenic, that Ben is a

81

top scholar, and that his parents are abusive. All of these descriptions were made about Ben and

his family at one point or another. These thin descriptions translate into labels one makes of him

or herself such as “I am from a crazy family” or “my life won’t amount to much.” They are

problem-saturated and obscure the strengths that underlie Ben’s unique human character

(Goldenberg & Goldenberg, 2012).

In contrast, thick descriptions speak to the unique nature of a person’s history and

identity; they are rich, deep, and diverse. Furthermore, thick descriptions are laden with

commentary and interpretation. For Ben, such a description might have been, “I come from a

long line of Navajo craftsmen, particularly silversmiths. I have learned from my father, who has

had his troubles, yet has been a successful silver smith in recent years. I have a wide variety of

skills drawn from my experiences growing up in a Navajo community as well as the time I have

spent learning in a boarding school. I have been able to navigate two complex social worlds

between school and home.” Parallel with this concept is the conversation around whether an

imaginary companion is an indicator of health or pathology, one might consider the title of

“hallucination” for an imaginary companion to be a “thin” description while the title “imaginary

companion” has thickness and weight to it: meaning that is missed if it were to be understood as

a hallucination.

Regarding his imaginary companion, Ben might have crafted a thick narrative description

of the development of his companion if given a chance, describing the role of the companion in

his life and the relationship of Ben’s companion to Ben’s unique sense of himself. His therapist

would likely be interested in the bigger picture of the role that this companion has had in Ben’s

life, considering the qualities of the relationships that other’s have with Ben and taking note of

82

the loss Ben experiences once his companion is gone. Ben’s sense of loss, exemplified when he

said,

“I had a friend who taught me things, who I had fun with. He was always with me wherever I went. We shared our feelings and fears and everything together. When I felt low I used to go out by myself, used to talk to myself and it seemed like it would talk back to me. I don’t feel that way anymore. It feels like I’ve lost somebody, part of me or something like that – like a real close friend, someone you’ve had since you were a baby, like a brother”

(Freeman, Epston & Lobovitz, 1997, p. 165).

It is also within this sense of loss that it is clear that Ben has lost something very valuable to him

that has been a part of his life for a long time. Developing at the age of two and staying with him

through the loss of his siblings, his uncle, boarding school, and stress at home, the companion

had taken up protecting Ben in times of crisis. For Ben to give voice to the narrative of this

friend could mean giving voice to parts of his autobiography that have been obscured by the

dominant narratives of his life (Freeman, Epston & Lobovitz, 1997). Furthermore, an

examination of the companion’s role and impact on Ben’s life through re-authoring

conversations could open space for a discussion of what is missing in Ben’s life when the

companion is not there; an important element of the work with Ben around loss. A narrative

therapist might ask Ben,

“What does your companion see in you?”

“What knowledge might someone who has known you so long have about you that is still unknown to others, even your parents?”

“We say: ‘Beauty is in the eye of the beholder.’ What would this beholder most want us to know about you and where you are in your life right now?”

It might be helpful for Ben to have a therapist who could explore the companion through the lens

of “one who saw early on that Ben needed help, and did not hold back from getting involved

83

with him,” a thick description which lends itself to meaning making as well as recognition of the

importance of attachment as well as the presence of trauma and loss.

Within re-authoring, the narrative therapist too remains curious and engaged with the

alternative stories that are possible and is committed to thickening the narrative of the client’s

life. In this way, the therapist and client co-create a larger volume of stories; opening space for

interpretation and meaning making that was not there when stories about the person were built

on thin, fact-based narratives.

Caveats

As was mentioned earlier, one of the aspects that proves challenging in using a case such

as Ben’s is that the narrative treatment is hypothetical. We do not know what Ben would have

brought up during his initial visit when approached with narrative questions. However, we can

imagine different directions that a narrative therapist might take, given certain responses from

Ben and his family.

With that said, it is important to notice in Ben’s story that the problem he came in with

was not that of the imaginary companion and the evil Spaniard spirit. Ben presented as mute and

dysregulated after a traumatic incident, one of a long line of traumatic incidents within his

relationships with his parents. The reader is reminded of the findings of Koverola and Foy (1993)

in that children may avoid narrating the significant traumatic events that have occurred in their

past given the way that they evoke aversive emotional reactions associated with the trauma. For

Ben, this might have meant total silence, as the acuity of his experience was intense. Or Ben

could have been silent due to the various conversations going on in his head between Ben, the

companion spirit and the evil spirit. The narrative therapist would remain open to multiple

possibilities and multiple interpretations of Ben’s presentation.

84

Through thoughtfully engaging Ben’s parents, as well as through offering mediums of

expression that Ben could relate to, his therapist could show that he/she is curious, interested and

very partial to Ben and is open to the stories the family has to tell. Additionally, given the recent

trauma Ben had experienced, it may be important for the therapist to work separately with Ben

and his parents, providing education to Ben’s parents about trauma and helping them to

understand Ben’s reactions and needs. Gradually, the work could come together, however it may

be best for this to be something for the family to work toward, when Ben shows through

communication that he is ready and feels safe engaging in therapy with his family. Finally, the

therapist would want to be careful in engaging Ben’s father. It is likely that the Ben’s

relationship with his father would be a source of fear and confusion given that Ben’s father had

recently been so violent and unpredictable, putting Ben and his family in great danger. The

narrative therapist would be interested in hearing the stories about Ben which capture the whole

person that he is as they work to create an externalization of the presenting problems.

Winnicottian Perspective

Winnicott’s theories date back much further than those of narrative therapy and were

influenced by the traditions and ideas of early psychoanalytic thinkers as well as psychoanalysts

practicing from the perspective of object relations (Mitchell & Black, 1995). While Winnicott

did not discuss the world in the postmodern terms that are used by narrative therapists, he did

take interest in the early caregiving environment and the impact of this environment on a

person’s unique intra psychic and relational development. Winnicott was interested in the way a

patient’s early caregiving relationships, and the environment wherein they were formed, manifest

as internal “object” representations; an internal world of relationships that Winnicott believed to

be more powerful than one’s relationships with actual people (Flannagan, 2008). In this way, an

85

individual’s family culture and childhood were central areas of inquiry and study for Winnicott

and the practice of addressing unmet needs and traumas from this time of life was central to

Winnicott’s treatment approach. Less central was an overt discussion of external cultural

influence on the caregiving environment as would be explored through a narrative approach.

Winnicott’s approach recognizes assessment and formulation as important treatment tools and

encourages therapists to provide a safe and holding context with which to support their patients,

while exploring new ways of relating in caregiving relationships. The following section will

address how a clinician practicing from a Winnicottian perspective might approach the

formulation and treatment of Ben within the context of psychotherapy

The Stance of the Therapist: Assessment and Treatment Considerations

Within any relational therapy, assessments are made at both intrapersonal and intra

psychic levels (Curtis & Hirsch, 2011). Winnicott (1969) discussed the nuances of assessment

from his perspectives when he wrote, “in our work, it is necessary for us to be concerned with

the development and the establishment of the capacity to use objects and to recognize a patient’s

inability to use objects, where this is a fact” (p. 711). Winnicott’s three concepts of the good

enough mother, transitional objects, and aloneness are especially helpful foundational principles

for considering the phenomenon of imaginary companions in adolescents with histories of

attachment trauma. The therapist could use each of these concepts to guide assessment and

treatment, paying close attention to the relational space between his or herself and Ben. Careful

attunement to the transference and countertransference dynamics between Ben and his therapist

would provide a rich sources of information about Ben’s relational framework and needs

(Mitchell & Black, 1995). Curtis and Hirsh (2011) emphasize that the therapist would want to

pay attention to, “the interactional styles, styles of coping and defenses, the range of emotions,

86

cognitive abilities, feelings about oneself and others, and conflicts and inhibitions that may block

the patient from achieving his or her goals” (p. 80). Furthermore, the therapist’s use of him or

herself in treatment would be informed by his or her perception of Ben’s unmet developmental

needs, communicated through the therapist’s observation of Ben’s interactional style in therapy

as trust builds.

Winncott believed that effective treatment is possible because of a therapeutic

relationship that mirrors the role of the good-enough mother, facilitating regression, awakening

old developmental needs, and supporting the client to use the therapeutic relationship to learn a

new style of relating to provide for what was missing in childhood (Mitchell & Black, 1995).

Frank (1999) describes this dynamic further when she writes,

while old object relations emerge in the transference, there is potential in the analytic space for new object experiences. For the new object experience to take hold, the treatment must engage the tenacity of the old. These experiences emerge in response to analytic attitudes which evoke words, responses, and reactions in the patient that had previously not been spoken (p. 434) Hence treatment would involve paying attention to enactments, impasses, ruptures and feelings

of closeness and distance, looking for opportunities to create new and holding object

experiences, addressing repair after ruptures, and engaging in styles of relating which address

unmet developmental needs and/or developmental traumas.

In Winnicott’s 1969 publication, The Use of an Object, he discussed the importance of

allowing a client’s transference feelings to develop overtime, emphasizing a lesson he had

learned through years of therapy about the importance of exercising restraint with psychoanalytic

interpretations and allowing the client’s trust for the therapist to grow by staying with the client

and listening well. Winnioctt (1969) said, “if only we wait, the patient arrives at understanding

creatively and with great joy” ( p. 711). Of the patient and the process he said, “I think I interpret

87

mainly to let the patient know the limits of my understanding. The principal is that it is the

patient and only the patient who has the answers” (p. 711). With these principals of the

therapist’s stance in mind, the reader is invited to think through assessment and treatment,

considering Ben’s presentation in the way that Winnicott might have, with special attention to

Winnicott’s concepts of good-enough mothering, transitional objects and aloneness.

The Good Enough Mother and The Holding Environment

Within the discussion section of Proskauer, Barsh and Johnson’s (1980) case study of

Ben, the authors write about Ben’s experience with parenting and the development of personal

resilience, writing,

Reared by a hostile controlling mother, an alternatively violent and passive schizophrenic father, and a bureaucratic boarding school system designed by an alien culture, Ben managed to create within himself a single male object which provided all that he needed: a friend, teacher, conscious, protector, source of personal power, and substitute for risky adolescent rebellion and experimentation

(p. 168).

Ben’s early life was challenging and chaotic. Ben’s parents struggled with managing the care of

their children, marriage and personal issues. In addition, the family experienced the traumatic

loss of four children who did not thrive beyond early infancy.

Reading the case of Ben, I found myself wondering what Ben’s parents were like on a

day-to-day basis, if they were tuned into Ben’s distress, attended to his physical and emotional

needs, and whether Ben had experiences of feeling known or seen by his caregivers. I wondered

what they had internalized about their Navajo identity amidst a white-man’s world throughout

their childhoods within the context of community, family and school. What kind of traumas were

they grappling within while parenting and what kind of intergenerational transmission of trauma

did Ben experience? Furthermore, there is reason to believe that some of Ben’s basic needs may

not have been met; the authors record that at least one of Ben’s brothers died of malnutrition.

88

Ben’s family was subject to systemic issues of poverty, lack of access to culturally congruent

education systems, discrimination from the perspective of dominant society, issues of oppression

based on the marginalization and trauma of Native American culture with regards to spirituality,

land ownership, traditions, autonomy and a host of other issues which created an environment of

struggle and strife in their daily life. All of these experiences were at play in shaping the context

of holding provided by Ben’s parents and his environment as well as the subsequent mapping of

these experience onto Ben’s psyche. Ben took his early caregiving experiences and the context of

his upbringing with him as he navigated each and every experience thereafter. While Winnicott

did not explicitly engage in cultural formulation, a formation from the a social work perspective

must account for the person within their environment, considering layers of marginalization,

oppression and privilege from the perspective of the person within the sociopolitical context.

Additionally, the clinician would also need to situate themselves in the sociopolitical context,

considering the identities they bring to the treatment and the ways in which their shared and

unshared identities may impact the treatment, broaching the topic as is appropriate to the

treatment (DeVines, 2007). With what we know about Ben from what is written in the case, I

will use Winnicott’s ideas to parse out some of the intrapsychic and interpersonal dynamics that

are central to Ben’s presentation.

For Winnicott, a mother must suspend her own subjectivity when she has a newborn,

immersing herself in catering to the infant’s needs and supporting the infant to develop their own

subjectivity. In this way, infants learn to believe in their own subjective omnipotence, meaning

that the infant believes that his own wishes create that which he desires. For instance, the infant

believes that his wish for nourishment creates the breast of his mother. For subjective

omnipotence to happen, the parent must be there when needed and give space when not needed

89

The parent feeds, soothes and attends to the infant on demand and, in this way, the optimal

holding environment is developed. The protective space of this environment allows the baby’s

spontaneity to exist and for them to express needs without hesitation. The holding environment is

crucial for development of the infant’s capacity to self-soothe. Eventually, the mother is unable

to maintain this suspended attention and the infant begins to learn that the mother has her own

subjectivity; learning about both objectivity and subjectivity is a new developmental task and

one that the infant is prepared for if they had a good-enough environment to begin with in

infancy (Mitchell & Black, 1995).

A therapist oriented to the ideas of Winnicott would wonder about the caregiving

environment in his early years vis-à-vis the information they receive from his adolescent

presentation. The environment of Ben’s home was likely one of unpredictability and angst, given

his father’s periodic angry and violent outbursts and his mother’s hostile and controlling

demeanor. According to Winnicott, a child’s emotional development is compromised when an

infant’s caregiver is unable to provide a good-enough holding environment. Instead of being

held, the child experiences themselves as an impingement and are forced to focus on aspects of

the external world at the cost of their internal world. If this failure on the part of the caregivers is

chronic, this can result in a split within the internal world of the infant through which they

develop a disconnect between their true self and false self; inhibition of the development of their

own subjectivity. Winicott believed that the environment the mother provides indicates the

infant’s ability to develop a sense realness (Mitchell & Black, 1995).Upon first meeting Ben, it is

likely that a therapist using an object relations perspective would take note of the split within

Ben’s psyche, represented by his experience of parts within himself: the companion spirit, the

90

evil Spaniard girl, and the part of Ben that was suspended and speechless when he came to the

mental health unit.

From a trauma perspective, Ben’s psychic split is indicative of a response to chaos and

trauma (i.e. witnessing his father attack his mother and then come after him) and provides

information about Ben’s developmental capacity to tolerate overwhelming incidents within

caregiving relationships. Drawing information from this intrapsychic response to a recent

traumatic incident, it is reasonable to believe that Ben was subject to trauma early in

development and was displaying what Allen (2011) describes as the “enduring and adverse

impact of extremely stressful events” (p. 4). As the reader may recall, trauma research indicates

that the experience of feeling terrified and alone can leave lasting and damaging psychic effects,

rendering individuals defenseless of cognitive structures and the capacity for emotional

regulation that would be present in the minds and bodies of non-traumatized individuals. Ben

may have been lacking the cognitive structures needed to manage his own subjectivity and hold

onto psychological organization because the very people that he needed to depend upon to foster

that organization and holding were also the very people that could not provide an environment of

predictability, trust and safety (Allen, 2011).

When a child does not experience caretaking, Winncott believed that the their primitive

ego works on learning to take care of the self and, out of this, the false self develops. Of this

false self, Winnicott says, “when a False Self becomes organized in an individual who has a high

intellectual potential there is a very strong tendency for the mind to become the location of the

False Self, and in this case there develops a dissociation between intellectual activity and

psychosomatic existence” and, “the world may observe academic success of a high degree, and

may find it hard to believe in the very real distress of the individual concerned, who feels

91

‘phony’ the more he or she is successful” (Winnicott, 1969, p. 3). Ben experienced a mother who

was preoccupied with a great deal of stress when Ben was young and Ben learned ways to both

take care of himself and remain loveable by his mother. Through this focus on the expectations

and desires of the external world, Ben learned to conform to the environment around him,

developing a false self manifested by academic success, a clean cut image and his way of staying

out of trouble. Ben is likely to have missed important developmental support from his caregivers

in cultivating soothing objects for his internal world. Ben may have learned that merit and

success could bring him a sense of recognition and a love-like feeling that was missing while he

was at school, developing a false self in order to get those needs met. He may also have learned

that loyalty and consistency were important values to assimilate in order to remain close to his

needed caregivers; that returning to his family could bring him a sense of acceptance within his

family and culture.

It appears that Ben recognized that the companion spirit and evil Spaniard girl within his

psyche were in conflict and that he desired a sense of wholeness within himself; perhaps a

connection to what Winnicott would call his true self. Winnicott would be curious about the

representations of spirit and companion parts as well as the role that Ben’s false self and these

parts play within his psyche. Perhaps, the parts are internalizations of the needed parts of his

caregivers (love, affection, and guidance) as well the bad parts of his caregivers (shame,

aggression and confusion). Perhaps the conflict between these parts mirrors the conflict that Ben

witnessed between family members, maintained in Ben’s psyche as a way of preserving the

holding environment that he knew. I imagine that the way that Ben felt in the midst of psychic

conflict was much like the way he felt within his family as a young child; caught between the

desire to be held and loved and the experience of feeling controlled and afraid. Finally, it’s

92

possible that these parts represented part of Ben’s identity that he felt an unconscious need to

split off as a way to conform to the dominant culture around him. In this way, an experience of

racial and ethnic shame could have led to the development of a companion spirit that is held

internally rather than expressed externally, something that Ben could have learned to do from

witnessing his parents’ relationship to their culture, race and ethnicity and from navigating his

own experience in a white-man’s word that encouraged conformity and certain behavioral and

academic standards. Was Ben encouraged to reject aspects of himself to fit with the status quo?

Was Ben encouraged to reject the white-man’s world at school or when he returned home? Were

his parents encouraged to reject parts of themselves? What was it like to be a young Navajo man

in the 1980’s? How did his community embrace culture pride and celebration as well as cultural

oppression and shame?

Ben may have gained an unconscious way of knowing that, if he could internalize a

loving image of his caregivers, Ben could survive the chaos within the environment and take care

of himself thereafter. In Winnicottian terms, the internalization of the imaginary companion,

then, could be understood as a resilient response to a not good enough holding environment. His

psychic struggle to destroy the evil Spaniard girl could have represented an unconscious desire to

destroy the bad parts of his parents contained within himself: the controlling, the violence, the

unpredictability. By asking Ben’s family to go to Spider Rock with Ben in order to destroy the

evil Spaniard girl, Ben was unconsciously asking his family to attune to him, to show up when

he asked them to do something, and commit themselves to a process of change and growth; a

collective family decision to destroy the badness inherent within the family system and the

culture surrounding them. Perhaps Ben had an unconscious wish to no longer carry the pain of

these dynamics on his own, desiring to elicit a collective response to change them .

93

Grief and loss were also a part of Ben’s early experiences. Ben lost two younger brothers

shortly after their birth, when Ben was two and five years old respectively. From Ben’s narrative,

as told through the voice of his companion, it is apparent that Ben internalized a sense of badness

and a fear of intimacy early in his life. Proskauer, Barsh and Johnson (1980) recount of Ben’s

narrative through the voice of his companion, “If a friend becomes too close, that friend dies,

because of the evil one; that’s how Ben’s two brother’s died. Ben felt that evil came from him.

Ben’s brother became very close to him, then the evil one made his brother die” (p. 164). One

can imagine a very young Ben, scared and craving attention and security, bonding to his new

born brothers, loving them as best as he knew how and witnessing their short life and early

death. How confusing and scary it must have been for Ben to make sense of this time, identifying

at least one of his brothers’ deaths as his personal fault and the result of badness within him.

How understandable then that Ben would want to destroy bad parts of himself, risking that good

and loving parts, which had been internalized and represented by the spirit companion, leave

along with the evil spirit. Proskauer, Barsh and Johnson (1980) write, “the male spirit companion

declared that after he had departed and the evil one had perished, Ben would no longer believe

that people die if they get too close to him and he would be able to make friends” (p. 164). It

seems that Ben believed he contained the omnipotence and power to hurt and even destroy those

around him. Ben may have seen himself as responsible for his mother’s anger and hostility, his

father’s violence and unpredictability, and the loss of those close to him. It is evident through the

narrative of Ben’s companion that a part of him believed that he was powerful enough to destroy

anyone he loved.

94

Treatment Considering Attachment Dynamics

Within the case of Ben, a Winnicott-oriented therapist would be concerned with the

development of Ben’s subjectivity and the ways in which he has adapted to the world of objects.

In treatment, Winnicott viewed the therapist as the agent to heal the split resulting from parenting

that was not good enough, using him or herself to attune to the client and provide a space of

holding in the therapy that could allow the client to “be” and experiencing nurturing and safety in

such a way that the true self could emerge. It is likely that this therapy would facilitate regression

in Ben, a way to experience holding for unmet developmental needs. Within the case of Ben, it is

apparent that his experience in hypnotherapy facilitated a regression to unmet needs for

subjective omnipotence. In the subjective omnipotent place, the child creates the object for his

own needs and then destroys it, taking it over (Mitchell & Black, 1995). Returning to the idea

that Ben asked his family to join him in the destruction of his spirit companions, for Winnicott,

the important task here would be for Ben to become aware of the caregiver that survives this

kind of destruction, whether it be his parents through accompanying him in this journey or his

therapist as Ben works through these changes in therapy. Through this process of holding and

survival, Ben could create a sense of external reality and realize that he no longer needs to be

ruthless. A careful holding environment in therapy could allow Ben to explore unmet needs,

asking for attunement from his therapist and support in experiencing regression while feeling

held, in order to achieve unmet developmental needs.

The reader may remember the early research conducted by psychoanalysts which

surmised that when imaginary companions were encountered in the play of older school aged

children and adolescents they were believed to be an indicator of a child’s underdeveloped

capacity to blend fragmented parts of themselves (Gupta & Desai, 2006; Friedberg, 1995) or a

95

way to defensively split off memory, feeling or part of the self (Taylor, 1999). The reader is also

reminded of the trauma research especially Allen’s (2011) writing about attachment trauma in

that, “attachment trauma damages the safety-regulating system and undermines the traumatized

person’s capacity to use relationships to establish feelings of security” (p. 22). Furthermore, the

thoughts of LaMothe (1999) regarding the adaptation of the self to attachment trauma, “there is a

horrifying void of an obliged, trustworthy, and faithful other an in this void there is consequently

the absence of psychological organization and the very structures of subjectivity […] these

experiences cannot be organized because the very symbols which human beings depend upon for

psychological organization cannot represent the very absence of obligation, trust and fidelity” (p.

344). Allen (2011) encourages readers to consider fantasy production, such as that represented

by an imaginary companion, as a mechanism of defensive coping called dissociation. This

research is summarized again here because the elements which serve as foundational for

Winnicott’s theories of good-enough mothering, are closely woven into these ideas about trauma,

self-hood, fragmentation, coping and development and must be held as such for a therapist

operating from an understanding of Winnicott and object relations.

Transitional Object

From a Winnicottian perspective, the comforting split off imaginary companion Ben

experienced could be viewed as part of the self that was used as a transitional object, kept

internally to manage transition and distance from home, family and culture throughout his life.

Returning to Winnicott’s idea of subjective omnipotence, a transitional object is viewed by

Winnioctt to be an auxiliary in the developmental shift between total dependence on the mother

(subjective omnipotence) to an “objective reality,” allowing the infant to both attach in early

attachment relationships as well as separate from them maturely (Mitchell & Black, 1995, p.

96

127). Referring back to the writing of Lane and Tolman (2007), good enough mothering “lays

the groundwork for integration of the mother object in each person’s inner world, a piece of that

psychological apparatus that eventually allows one, through mental functioning to perform thee

care-giving activities for one’s self” (p. 41). Winnicott believed that a transitional object, such as

a stuffed animal or blanket could stand in for the mother as a self-soothing object, representing

parts of the mother and parts of the child during this transitional time of development (Goldstein,

2001). Ben’s spirit companion reportedly first came to him at the age of two, which the

psychological literature indicates is developmentally normal. However, Ben’s companion seems

to have come to support him in more than the typical developmental achievements. Ben’s

companion stepped in during times of chaos, crisis and transition. Ben is quoted as saying,

I had a friend who taught me things, who I had fun with. He was always with me wherever I went. We shared our feelings and fears and everything together. When I felt low I used to go out by myself, used to talk to myself and it seemed like it would talk back to me. I don’t feel that way anymore. It feels like I’ve lost somebody, part of me or something like that – like a real close friend, someone you’ve had since you were a baby, like a brother

(Proskauer, Barsh & Johnson, 1980, p. 165).

This companion was an internalization of the loving and comforting caregiver that he so needed

to deal with the isolation of early childhood in the midst of a “hostile” and “controlling” mother

who, in her own way, was working to manage emotional pain and her own history of multiple

macro and micro traumas (Proskauer, Barsh & Johnson, 1980, p. 168).

Winnicott’s concept of transitional objects described them as concrete objects, describing

them as especially helpful when depressive moods arise for children (Winnicott, 1971).

Physically, transitional objects such as these can provide soothing touch, softness, and something

to cuddle when the mother is not there. Psychologically, the object represents that which is not

there, namely the mother. Winnicott believed that, if children get stuck in relative dependence,

97

they will have problems with separation and individuation later in life. In this vein, the reader is

reminded of Winnicott’s description of a 7-year-old boy who identified with string in his

mother’s absence. In spite of therapy to address this issue and intervention on the part of the

boy’s parents, he was never able to manage separation from his mother and leave the home for a

life of independence (Winnicott, 1971). The question arises here for this author, what makes

Ben’s experience different than this seven year old boy who could never leave home? What

enabled Ben to develop into a mature and responsible young person with goals and aspirations?

Winnicott might say that Ben created a comforting image of a soothing caregiver in spite

of the absence of such a caregiver. Hence, when Ben left home, this soothing object remained

inside of him, supporting him with transitions and losses much as it had in his early childhood.

Ben created something for himself that could not leave him until he entered therapy at the mental

health clinic. During the course of therapy, Ben lost the spirit companions, comparing this loss to

the pain of losing a close friend or a brother; loss that Ben knows well about. The loss of his

companion seemed to open up a deeper well of unresolved grief including the loss of his siblings

and his uncle. It is important to consider whether the loss Ben of his imaginary friends was a

goal of treatment or merely a byproduct. If it was a goal of treatment, perhaps Ben learned that it

was shameful to have imaginary companions and experienced judgment from those he told his

about his companion. If the loss was a byproduct of treatment, then perhaps Ben had engaged in

a healing experience of regression and treatment for the losses and inadequacies of his early

childhood. In any event, this companion is likely to have represented a transitional space for Ben,

a support in times of loneliness and crisis, that was dependably with him time and time again

when he was in need. This internalized object provided the comfort and soothing that he needed

98

to come back to himself and confront the world feeling less alone (Proskauer, Barsh and

Johnson, 1980).

Aloneness

In Winnicott’s (1958) writing about aloneness, he drew an important distinction between

the act of spending time alone and the capacity to be alone. For Winnicott, the capacity to be

alone required an experience in infancy wherein the infant experiences themselves as able to be

alone in the presence of another person. Through this kind of aloneness experience, the infant

can internalize the good object of the soothing caregiver that sat with them. The good-enough

mother in this scenario needs to allow the infant the time and space to explore the world

independently and then satisfy the infant’s instinctual gratifications by making themselves

available to meet the infants needs as they arise.

As stated earlier, there is reason to believe that Ben’s parents were not “good enough” to

meet his dependency needs in early childhood. They raised Ben amidst the loss of four other

children in infancy and were likely unable to meet Ben’s needs due to preoccupation with their

own emotional needs and stressors within the environment. Ben’s spirit companion was

especially helpful to him during times when he felt most alone; with the support of his

companion, Ben didn’t have to feel the depth of loneliness. In this way, Ben could self-soothe

without the need to depend on an external object. However, later in life, Ben found it difficult to

experience intimacy and connection in relationships. Thus, the adaptive coping skills of relying

on his imaginary companion did not work for him later in life when he desired real and tangible

human connection. In this way, Ben was able to utilize therapy to learn about his use of the

imaginary companion and to express desire for more connection, finding this both through

99

therapy and later through the relationship with his girlfriend (Proskauer, Barsh and Johnson,

1980).

Winnicott Treatment Caveats and Conclusions

Winnicott’s concept of the good enough mother leaves room for human messiness on the part

of the mother, sharing that a mother need not be perfect, only good-enough to develop an internal

world of subjectivity that allows the infant to experience human life as real an meaningful

(Mitchell & Black, 1995). However, when the environment is not good enough, the false self

develops. Ben has been a consistent academic performer in school, doing well to satisfy the

requirements and stay out of trouble within the school setting. He was also careful socially,

intent to remain away from alcohol and drugs despite temptation. Perhaps these accommodations

represent the false self Ben developed to deal with a chaotic upbringing, a lack of self-soothing

interjects, and a need and desire to be seen and appreciated by others.

Because of the connection between imaginary companions and attachment trauma (vis-a-

vis fragmentation) a therapist would need to be particularly delicate when addressing previously

unmet attachment needs. The self is vulnerable to fragmentation and the role of the therapist is

one of re-parenting and holding. To traumatize the individual during this process by failure or

misattunement, repeated and unaddressed enactments, could be to do further damage. The

therapist would need to be careful with how quickly the role of the imaginary companion

diminishes, understanding the important role the companion has played in Ben’s life and

allowing a safe and trusting attachment to form with the therapist so that slow growth,

interpersonally and intrapsychially, could take place. Finally, perhaps the goal of therapy would

not be to rid Ben of his companion experience but to support him in learning to build intimate

and safe in-person relationships with others.

100

Synthesis and Implications for Social Work Practice

Since psychotherapists are the intended audience of this thesis, one goal of this thesis is

to create awareness in clinicians who may treat adolescents who present with imaginary

companions and a history of attachment trauma. This thesis has presented the theories of

narrative therapy and Winnicott in order to offer clinicians lenses through which to view a

complicated presentation. Most importantly, this thesis encourages practitioners to open their

minds to the idea that the imaginary companions of traumatized adolescents can be viewed as a

response of resilience and adaptation to an otherwise troubling and upsetting world. Perspectives

on trauma help the reader to see how attachment trauma can lead to dissociative phenomenon as

well as a fragmentation of the self; primary survival responses to this particular kind of trauma.

The dynamics of attachment trauma and personality development are complex and must be

treated carefully and with an emphasis on how the person had adapted to the world as they have

known it. In this way, psychotherapists are encouraged to engage in a curious assessment, using

the aforementioned perspectives as guides to understanding how imaginary companions may

have come to creation in the lives of their clients. The willingness of the therapist to understand

the companion in a positive light, as the child experiences them, is necessary for developing an

understanding of the resilience that helps bring people through the most terrible kinds of trauma.

An additional goal of this research is to thoughtfully contribute to the limited body of

research that has been done on the phenomena within psychoanalytic literature. As the

phenomenon chapter explained, there has been little research conducted in the field, most of it is

dated, and a great deal of the research indicates that the phenomenon is pathological in nature.

This thesis has offered an alternate perspective, shedding light on the resilient function of

imaginary companions, while also holding in the balance concerns for treatment and treatment

101

outcomes. In all, this thesis endeavors to enhance the research that in the field and challenge

early findings by offering ideas and perspectives stemming from more nuanced theories,

understanding and approaches.

Winicott talks about failures within the holding environment without recognizing the

failures within the system which provide a context for families and holding environment.

Readers are also encouraged to consider the systemic issues relevant to their adolescent clients

with trauma histories and imaginary companions. Oppression, systemic racism and

discrimination were the backdrop his Ben’s family’s day-to-day reality. A therapist, social

worker or otherwise, would be encouraged to address unmet developmental needs within the

family system as well as unmet needs of personal agency, self-determination, equality, power,

etc. Furthermore, Ben’s experiences should be considered a call to action and important

indicators of the need for justice and equality for the Navajo people. As social workers, we have

an ethical responsibility to live out the values of our profession, values of service, social justice,

dignity and worth of the person, the importance of human relationships, integrity and

competence. These values must not be lost amongst the important clinical work that social

workers engage in with their clients; the sociopolitical context that we all live in is personally

relevant and should be explored both in therapy and through social action (National Association

of Social Workers, 1999)

The findings within the literature review of this thesis indicate that, while there has been

some research done on this important phenomenon, there is still much research to be done. The

concerning cycles of abuse that Ben experienced are the reality for many children across the

country and throughout the world. As long as poverty and inequality exist, access to many of the

sociopolitical factors that were salient in Ben’s case will enable the system of power and abuse

102

that exist within families and the wider society. Social work and the associated arms of policy

and research are encouraged to continue to address the macro systems within our society and to

continue to trouble and fight for the human rights of our clients.

Conclusions

Throughout the present study, this author has thoughtfully drawn connections between

the phenomenon and theories of psychology and treatment, crafting an integrative approach to

the phenomenon that takes into account multiple perspectives from which to understand the

phenomenon. However, this study is not without its limitations. This author’s application of

theory was to an older case that was not written with this study in mind. This study would have

been enhanced by the opportunity to work with this author’s personal case material and a client

that this writer knows intimately. Furthermore, looking at more than one case could have

enhanced the study, through perhaps, conducting interviews with clients experiencing the

phenomenon, their parents, their teachers and their therapists. Further research is recommended

in the area of imaginary companions of older children who have experienced attachment trauma

and this writer encourages researches to further explore the realms of resilience present within

the presentation of this phenomenon. May the thoughtful inquiry continue and the curious

inquirers always lend an ear to the legacy of personal and collective pain and power in the

creation of new and empowering narratives.

103

REFERENCES

Adamo, S.G. (2004). An adolescent and his imaginary companions: From quasi-delusional

constructs to creative imagination. Journal Of Child Psychotherapy, 30(3), 275-295.

Adams-Westcott, J., & Dobbins, C. (1997). Listen with your ‘hearts ears’ and other ways young

people can escape the effects of sexual abuse. In C. Smith & D. Nylund (Eds.), Narrative

therapies with children and adolescents (1-52). New York: The Guildford Press.

Allen, J. (2011). Traumatic relationships and serious mental disorders. New York : John Wiley

& Sons.

Allison, R. B., & Schwartz, T. (1980). Minds in many pieces. New York: Rawson, Wade.

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders

(4th ed. text rev.). Washington, DC: Author.

Bach, S. (1971). Notes on some imaginary companions. The Psychoanalytic Study of The Child,

26, 159-171.

Bender, L., Vogel, B.F. (1941). Imaginary companions of children. American Journal of

Orthopsychiatry, 11(1), 56-65.

Benson, R. M., & Pryor, D. B. (1973). When friends fall out: Developmental interference with

the function of some imaginary companions. Journal of the American Psychoanalytic

Association, 21, 457-473.

Blizard, R.A. (1997). The origins of dissociative identity disorder from an object relations and

attachment theory perspective. Progress in the Dissociative Disorders, 10(4), 223-229

Blizard, R.A. (2003). Disorganized attachment, development of dissociated self states, and a

relational approach to treatment. Journal of Trauma & Dissociation, 4(3), 27-50.

104

Briere, J., & Runtz, M. (1990). Augmenting Hopkins SCL scales to measure dissociative

symptoms: Data from two nonclinical samples. Journal of Personality Assessment, 55(1-

2), 376-379.

Brooks, M., & Knowles, D. (1982). Parents’ views of children’s imaginary companions. Child

Welfare, 61(1), 25-33.

Bruner, J. S. (1990). Acts of meaning. Cambridge, MA: Harvard University Press.

Center for Disease Control (2012). Child maltreatment: Definitions. Retrieved on 2 July, 2013.

Retrieved from

http://www.cdc.gov/violenceprevention/childmaltreatment/definitions.html

Compas, B.E., Hinden, B.R., & Gerhardt, C.A. (1995). Adolescent development: Pathways and

processes of risk and resilience. Annual Review of Psychology, 16, 265-293

Curtis, R.C. & Hirsch, I. (2011). Relational psychoanalytic psychotherapy. In Essential

psychotherapies: theory and practice. Eds. S.B. Messer & A.S. Gurman. Third Edition.

New York: The Guildford Press.

Davies, D. (2011). Child development: A practitioner's guide (2nd ed.). New York: Guilford.

Dell, P. (1982). Pathology: The original sin. Paper presented at the First International

Conference on Epistemology, Psychotherapy, and Psychopathology, Houston, TX.

Dierker, L.C., Davis, K.F., & Sanders, B. (1995). The imaginary companion phenomenon: An

analysis of personality correlates and developmental antecedents. Dissociation: Progress

In The Dissociative Disorders, 8(4), 220-228.

Dulwich Center (2013). Commonly asked questions about narrative approaches to therapy,

community work, and psychosocial support. Retrieved from

http://www.dulwichcentre.com.au/common-questions-narrative-therapy.html

105

Dunsany, L. (1926). Charwoman’s shadow. London: Putnam.

Durrant, M., & Kowalski, K. (1990). Overcoming the effects of sexual abuse: Developing a self-

perception of competence. In M. Durrant & M. White (Eds.), Ideas For Therapy With

Sexual Abuse. 65-110. Adelaide, South Australia: Dulwich Center Publications.

Fein, G.G. (1975). A transformational analysis of pretending. Developmental Psychology, 11,

291-296.

Fibush, R. (1998). Children’s recollections of traumatic and nontraumatic events. Developmental

and Psychopathology, 10(4), 669-716

Fivush, R., & Haden, C. A. (1997). Narrating and representing experience: Preschoolers’

developing autobiographical accounts. In P. W. Van den Broek, P. Bauer, & T. Bourg

(Eds.), Developmental spans in event comprehension and representation. Bridging

fictional and actual events (pp. 169–198). Mahwah, NJ: Erlbaum.

Flanagan, L.M. (2008). Object relations theory. In J. Berzoff, L.M. Flanagan, and P. Hertz

(Eds.), Inside out and outside in (pp. 189-204). New York: Aronson.

Foehrenbach, L., Celentano, C., Kirby, J., & Lane, R.C. (1997). Developmental determinants of

psychosomatic symptoms. In J. Finell (Ed.). Mind-body problems: Psychotherapy with

psychosomatic disorders (pp. 19-38). Lanham, MD US: Jason Aronson.

Frank, M.G. (1999). From monologue to dialogue: Intersubjective voices in the analytic space.

Smith College Studies of Social Work, 69(2), 429-443.

Freedman, J. & Combs, G. (1996). Narrative therapy: The social construction of preferred

realities. New York: Norton.

Freeman, J., Epston D., Lobovits, D. (1997). Playful approaches to serious problems: Narrative

therapy for children and their families. New York: W.W. Norton & Company.

106

Freud, A. (1936). The ego and the mechanisms of defense. The Writings of Anna Freud, 2:1-176.

Friedberg, R. (1995). Allegorical lives: Children and their imaginary companions. Child Study

Journal, 25(1), 1-21.

Giolas, M.H., & Sanders, B. (1992). Pain and suffering as a function of dissociation level and

instructional set. Dissociation: Progress in the Dissociative Disorders, 5I(4), 205-209.

Gleason, T. R. (2004). Imaginary companions and peer acceptance. International Journal of

Behavioral Development, 28, 204–209.

Goldenberg, H., & Goldenberg, I. (2013). Family therapy: An overview. California:

Brooks/Cole.

Goldstein, E.G. (2001). Object relations theory and self psychology in social work practice. New

York, NY, US: Free Press

Gupta, A., & Desai, N.G. (2006). Pathological fantasy friend phenomenon. International Journal

of Psychiatry In Clinical Practice, 10(2), 149-151.

Harriman, P.L. (1937) Some imaginary companions of older subjects. The American Journal of

Orthopsychiatry. 7(3), 368 – 370.

Harter, S., & Chao, C. (1992). The role of competence in children’s creation of imaginary

friends. Merrill-Palmer Quarterly, 38(3), 350-363.

Harvey, N.A. (1918). Imaginary playmates and other mental phenomena of children. State

Normal College. Ypsilanti, Michigan.

Hornstein, N.L., & Putnam, F.W. (1996). Abuse and the development of dissociative symptoms

and dissociative identity disorder. In C.R. Pfeffer (Ed.), Severe stress and mental

disturbance in children (pp. 449-473). Arlington, VA, US: American Psychiatric

Association

107

Hurlock , E. B. & Burstein , W. (1932) The Imaginary Playmate. J. Genetic Psychology, 41:

380-392.

Inuzuka, M., Satoh, Y., & Wada, K. (1991). The imaginary companion: A questionnaire.

Japanese Journal of Child and Adolescent Psychiatry, 32(1), 32-48.

Jersild, A.T. (1968) Child Psychology (6th ed.). Englewood Cliffs, NJ: Prentice-Hall.

Jersild, A.T., Jersild, C.L. and Markey, F.V. (1937) Children’s fears, dreams wishes, daydreams,

likes, dislikes, pleasant and unpleasant memories. Child Development, 12.

Kandall, E. (1997). Imaginary play companions and the children who know them. Unpublished

doctoral dissertation, The California School of Professional Psychology, Almeda.

Kisiel, C.L., & Lyons, J.S. (2001). Dissociation as a mediator of psychopathology among

sexually abused children and adolescents. The American Journal of Psychiatry, 158(7),

1034-1039.

Kita, E. (2012). Making it thinkable. In J. Berzoff (Ed.), Falling through the cracks:

Psychodynamic practice with vulnerable and oppressed populations. New York:

Colombia University Press.

Klein, B.R. (1985) A child’s imaginary companion: A transitional self. Clinical Social Work

Journal, 13, 272-282.

Koverola, C., Foy, D. (1993). Post traumatic stress disorder symptomatology in sexually abused

children: Implications for legal proceedings. Journal of Child Sexual Abuse, 2(4), 119-

128.

Liotti, G. (1999). Understanding the dissociative processes: The contribution of attachment

theory. Psychoanalytic Inquiry, 19(5), 757-783

108

Madsen, W.C. (2007). Collaborative therapy with multi-stressed families (2nd Ed.). New York,

NY, US: Guilford Press

Manosevitz, M., Prentice, N.M., & Wilson, F. (1973). Individual and family correlates of

imaginary companions in preschool children. Developmental Psychology, 8, 72-79.

McElroy, L.P. (1992). Early indicators of pathological dissociation in sexually abused children.

Child Abuse & Neglect, 16(6), 833-846.

McLeod, J. (2006). Narrative thinking and the emergence of postpsychological therapies.

Narrative Inquiry, 16(1), 201-210.

McLewin, L.A., & Muller, R.T. (2006). Childhood trauma, imaginary companions, and the

development of pathological dissociation. Aggression And Violent Behavior, 11(5). 531-

545

Meyer, J., & Tuber, S. (1989). Intrapsychic and behavioral correlates of the phenomenon of

imaginary companions in young children. Psychoanalytic Psychology, 6 (2), 151-68.

Mitchell, S. A., & Black, M. J. (1995). Freud and beyond: A history of modern psychoanalytic

thought. New York: Basic Books.

Monk, G. (1997). How narrative therapy works. In G. Monk, J. Winslade, K. Crocket, & D.

Epston, (Eds.), Narrative therapy in practice: The archaeology of hope (3-31). San

Francisco: Jossey-Boss.

Mossige, S., Jensen, T.K., Gulbrandsen, W., Reichelt, S., & Tjersland, O.A. (2005). Childrens’

narratives of sexual abuse: What characterizes them and how do they contribute to

meaning-making. Narrative Inquiry 15(2), 377-404.

Nagera, H. (1969). The imaginary companion: Its significance for ego development and conflict

resolution. Psychoanalytic Study of the Child, 24, 165-196.

109

National Association of Social Workers. (1999). Code of ethics of the National Association of

Social Workers. Washington, DC. NASW Press.

National Scientific Counsel on the Developing Child. (2009). Excessive stress disrupts the

architecture of the developing brain: Working paper 3. Center on the Developing Child:

Harvard University. Retrieved from www.developingchild.net

Nichols, M., & Schwartz, R. (2008). Essentials of family therapy (4th ed.). Needham Heights,

MA: Allyn & Bacon.

Ogawa, J.R., Sroufe, L., Weinfield, N.S., Carlson, E.A., & Egeland, B. (1997). Development and

the fragmented self: Longitudinal study of dissociative symptomatology in a nonclinical

sample. Development and Psychopathology, 9(4), 855-879

Pearson, D.D., Rouse, H.H., Doswell, S.S., Ainsworth, C.C., Dawson, O.O., Simms, K.K.,

Edwards, L., & Falconbridge, J.J. (2001). Prevalence of imaginary companions in a

normal child population. Child, Care, Health and Development, 27(1), 13-22

Piaget (1951). Play, dreams an imitation in childhood. New York: Norton.

Pica, M. (1999). The evolution of alter personality states in dissociative identity disorder.

Psychotherapy: Theory, Research, Practice, Training, 36(4), 404-415

Proskauer, S., Barsh, E.T., & Johnson, L.B. (1980) Imaginary companions and spiritual allies of

three Navajo adolescents. Journal of Psychological Anthropology, 3(2), 153-174

Putnam, F. W. (1997). Dissociation in children and adolescents. New York: The Guilford Press

Putnam, F.W., Guroff, J.J., Silberman, E.K., Barban, L., & Post, R.M. (1996). The clinical

phenomenology of multiple personality disorders: Review of 100 recent cases. Journal of

Clinical Psychiatry, 47, 285-293

110

Roberts, J. (1994). Tales & transformations: Stories in families and family therapy. New York:

W. W. Norton & Company.

Rosenwald, G.C., & Ochberg, R.L. (Eds.) (1992). Storied lives: The cultural politics of self-

understanding. New Haven: Yale University Press.

Roth, S. & Epston, D. (1996). Developing externalizing conversations: An exercise. Journal of

Systemic Therapies, 15(1), 5-12.

Sanders, B. (1992). The imaginary companion experience in multiple personality disorder.

Dissociation, 5(3), 159-162.

Sawa, T., Oae, H., Abiru, T., Ogawa, T., & Takahashi, T. (2004). Role of imaginary companion

in promoting the psychotherapeutic process. Psychiatry And Clinical Neuroscience,

58(2), 145-151.

Schore, A.N. (2001). The effects of early relational trauma on the right brain development, affect

regulation, and infant mental health. Infant Mental Health Journal, 22(1-2), 201-296.

Schultz, R., Braun, B.G., & Kluft, R.P. (1989). Multiple personality disorder: Phenomenology of

selected variables in comparison to major depression. Dissociation, 2, 45-51.

Seiffge-Krenke, I (1997). Imaginary companions in adolescence: Sign of a deficient or positive

development? Journal of adolescence, 20(2): 137-154.

Shirar, L. (1996). Dissociative children: Bridging the inner and outer worlds. New York, NY,

US: WW Norton & Co.

Singer, D. G., & Singer, J. L. (1990). The house of make-believe: Children’s play and developing

imagination. Cambridge, MA: Harvard University Press.

111

Smith, C. (1997). Comparing traditional therapies with narrative approaches. In C. Smith & D.

Nylund, (Eds.), Narrative therapies with children and adolescents (1-52). New York: The

Guildford Press.

Somers, J., Yawkey, T.D. (1984). Imaginary play companions: Contributions of creative and

intellectual abilities of young children. Journal of creative behavior, 18(2), 77-87.

Sperling, O. E. (1954). An Imaginary companion Representing a Pre-stage of the Superego. The

Psychoanalytic Study of the Child, 9: 252-258

Stone, E. (1988). Black sheep and kissing cousins. New York: Times Books.

Svedson, M. (1934). Children’s imaginary companions. Archives of Neurology & Psychiatry.

32: 985-999.

Taylor, M. (1999). Imaginary companions and the children who created them. New York, NY

US: Oxford University Press.

Taylor, M., & Carlson, S. M. (1997). The relation between individual differences in fantasy and

theory of mind. Child Development, 68, 436 – 455.

Taylor, M., Carlson, S.M., Maring, B., Gerow, L.E., & Charley, C. (2004) The characteristics

and correlates of high fantasy in school-aged children: Imaginary companions,

impersonation, and social understanding. Developmental psychology, 40, 1173- 1187.

Taylor, M., Hulette, A.C., & Dishion, T.J. (2010). Longitudinal outcomes of high-risk

adolescents with imaginary companions. Developmental Psychology, 46(6), 1632-1636.

Taylor, M., Shawber, A.B., Mannering, A.M. (1999). Imaginary companions and the children

who create them. New York, NY: Oxford University Press.

112

Tolman, M.D. & Lane, R.C. (2007). The role of the mother on the development of restricting

anorexia. In R.C. Lane & M.D. Tolman (Eds.), Psychodynamic perspectives on eating

disorders (pp. 176-197). Victoria, BC, Canada: Trafford.

Trionfi, G., & Reese, E. (2009). A good story: Children with imaginary companions create richer

narratives. Child Development, 80, 1301– 1313.

Trujillo, K., Lewis, D. O., Yeager, C. A., & Gidlow, B. (1996). Imaginary companions of school

boys and boys with Dissociative Identity Disorder/Multiple Personality Disorder: A

normal to pathological continuum. Child and Adolescent Psychiatric Clinics of North

America, 5, 375-391.

Walsh, N. (2012) Clinical views of family normality, health, and dysfunction: From a deficits to

a strengths perspective. In N. Walsh (Ed.), Normal family process: Growing diversity and

complexity (pp. 28-54).

Walton, M. D., & Brewer, C. L. (2002). The role of personal narrative in bringing children into

the moral discourse of their culture. Narrative Inquiry, 11(2), 307–334.

White, M. (2004). Folk psychology and narrative practices. In L.E. Angus & J. McLeod (Eds.),

The handbook of narrative and psychotherapy: Practice, theory and research (pp. 15–52).

Thousand Oaks, CA: Sage.

White, M. (2007). Maps of narrative practice. New York: W. W. Norton & Company.

White, M. (1995). Re-authoring lives: Interviews and essays. Adelaide, South Australia:

Dulwich Centre Publications.

White, M., Epston, D. (1990). Narrative meant to therapeutic ends. New York: Norton.

Wickes, F.G. (1996). The inner world of childhood: A study in analytic psychology. New York:

Appleton-Century.

113

Winnicott ,D.W. (1956). On transference. The International Journal or Psychoanalysis, 37, 386-

388

Winnicott, D.W. (1971). Play and reality. London: Tavistock Publications.

Winnicott, D. W. (1958). The capacity to be alone. In D. W. Winnicott, The maturational

processes and the facilitating environment: Studies in the theory of emotional

development (pp. 29-36). NY: International Universities Press, Inc.

Winnicott, D.W. (1969). The use of an object. International Journal of Psychoanalysis, 50, 711-

716.